Search

Multi-task learning solves multiple correlated tasks. However, conflicts may exist between them. In such circumstances, a single solution can rarely optimize all the tasks, leading to performance trade-offs. To arrive at a set of optimized yet well-distributed models that collectively embody different trade-offs in one algorithmic pass, this paper proposes to view Pareto multi-task learning through the lens of multi-task optimization. Multi-task learning is first cast as a multi-objective optimization problem, which is then decomposed into a diverse set of unconstrained scalar-valued subproblems. These subproblems are solved jointly using a novel multi-task gradient descent method, whose uniqueness lies in the iterative transfer of model parameters among the subproblems during the course of optimization. A theorem proving faster convergence through the inclusion of such transfers is presented. We investigate the proposed multi-task learning with multi-task optimization for solving various problem settings including image classification, scene understanding, and multi-target regression. Comprehensive experiments confirm that the proposed method significantly advances the state-of-the-art in discovering sets of Pareto-optimized models. Notably, on the large image dataset we tested on, namely NYUv2, the hypervolume convergence achieved by our method was found to be nearly two times faster than the next-best among the state-of-the-art.

The creation of large, diverse, high-quality robot manipulation datasets is an important stepping stone on the path toward more capable and robust robotic manipulation policies. However, creating such datasets is challenging: collecting robot manipulation data in diverse environments poses logistical and safety challenges and requires substantial investments in hardware and human labour. As a result, even the most general robot manipulation policies today are mostly trained on data collected in a small number of environments with limited scene and task diversity. In this work, we introduce DROID (Distributed Robot Interaction Dataset), a diverse robot manipulation dataset with 76k demonstration trajectories or 350 hours of interaction data, collected across 564 scenes and 84 tasks by 50 data collectors in North America, Asia, and Europe over the course of 12 months. We demonstrate that training with DROID leads to policies with higher performance and improved generalization ability. We open source the full dataset, policy learning code, and a detailed guide for reproducing our robot hardware setup., Comment: Project website: https://droid-dataset.github.io

Intelligent agents must be generalists, capable of quickly adapting to various tasks. In reinforcement learning (RL), model-based RL learns a dynamics model of the world, in principle enabling transfer to arbitrary reward functions through planning. However, autoregressive model rollouts suffer from compounding error, making model-based RL ineffective for long-horizon problems. Successor features offer an alternative by modeling a policy's long-term state occupancy, reducing policy evaluation under new tasks to linear reward regression. Yet, policy improvement with successor features can be challenging. This work proposes a novel class of models, i.e., generalized occupancy models (GOMs), that learn a distribution of successor features from a stationary dataset, along with a policy that acts to realize different successor features. These models can quickly select the optimal action for arbitrary new tasks. By directly modeling long-term outcomes in the dataset, GOMs avoid compounding error while enabling rapid transfer across reward functions. We present a practical instantiation of GOMs using diffusion models and show their efficacy as a new class of transferable models, both theoretically and empirically across various simulated robotics problems. Videos and code at https://weirdlabuw.github.io/gom/.

Imitation learning methods need significant human supervision to learn policies robust to changes in object poses, physical disturbances, and visual distractors. Reinforcement learning, on the other hand, can explore the environment autonomously to learn robust behaviors but may require impractical amounts of unsafe real-world data collection. To learn performant, robust policies without the burden of unsafe real-world data collection or extensive human supervision, we propose RialTo, a system for robustifying real-world imitation learning policies via reinforcement learning in "digital twin" simulation environments constructed on the fly from small amounts of real-world data. To enable this real-to-sim-to-real pipeline, RialTo proposes an easy-to-use interface for quickly scanning and constructing digital twins of real-world environments. We also introduce a novel "inverse distillation" procedure for bringing real-world demonstrations into simulated environments for efficient fine-tuning, with minimal human intervention and engineering required. We evaluate RialTo across a variety of robotic manipulation problems in the real world, such as robustly stacking dishes on a rack, placing books on a shelf, and six other tasks. RialTo increases (over 67%) in policy robustness without requiring extensive human data collection. Project website and videos at https://real-to-sim-to-real.github.io/RialTo, Comment: Project page: https://real-to-sim-to-real.github.io/RialTo

In recent years, significant progress has been made in the field of robotic reinforcement learning (RL), enabling methods that handle complex image observations, train in the real world, and incorporate auxiliary data, such as demonstrations and prior experience. However, despite these advances, robotic RL remains hard to use. It is acknowledged among practitioners that the particular implementation details of these algorithms are often just as important (if not more so) for performance as the choice of algorithm. We posit that a significant challenge to widespread adoption of robotic RL, as well as further development of robotic RL methods, is the comparative inaccessibility of such methods. To address this challenge, we developed a carefully implemented library containing a sample efficient off-policy deep RL method, together with methods for computing rewards and resetting the environment, a high-quality controller for a widely-adopted robot, and a number of challenging example tasks. We provide this library as a resource for the community, describe its design choices, and present experimental results. Perhaps surprisingly, we find that our implementation can achieve very efficient learning, acquiring policies for PCB board assembly, cable routing, and object relocation between 25 to 50 minutes of training per policy on average, improving over state-of-the-art results reported for similar tasks in the literature. These policies achieve perfect or near-perfect success rates, extreme robustness even under perturbations, and exhibit emergent recovery and correction behaviors. We hope that these promising results and our high-quality open-source implementation will provide a tool for the robotics community to facilitate further developments in robotic RL. Our code, documentation, and videos can be found at https://serl-robot.github.io, Comment: ICRA 2024

A network of nanomachines (NMs) can be used to build a target detection system for a variety of promising applications. They have the potential to detect toxic chemicals, infectious bacteria, and biomarkers of dangerous diseases such as cancer within the human body. Many diseases and health disorders can be detected early and efficiently treated in the future by utilizing these systems. To fully grasp the potential of these systems, mathematical analysis is required. This paper describes an analytical framework for modeling and analyzing the performance of target detection systems composed of multiple mobile nanomachines of varying sizes with passive/absorbing boundaries. We consider both direct contact detection, in which NMs must physically contact the target to detect it, and indirect sensing, in which NMs must detect the marker molecules emitted by the target. The detection performance of such systems is calculated for degradable and non-degradable targets, as well as mobile and stationary targets. The derived expressions provide various insights, such as the effect of NM density and target degradation on detection probability.

We consider imitation learning with access only to expert demonstrations, whose real-world application is often limited by covariate shift due to compounding errors during execution. We investigate the effectiveness of the Continuity-based Corrective Labels for Imitation Learning (CCIL) framework in mitigating this issue for real-world fine manipulation tasks. CCIL generates corrective labels by learning a locally continuous dynamics model from demonstrations to guide the agent back toward expert states. Through extensive experiments on peg insertion and fine grasping, we provide the first empirical validation that CCIL can significantly improve imitation learning performance despite discontinuities present in contact-rich manipulation. We find that: (1) real-world manipulation exhibits sufficient local smoothness to apply CCIL, (2) generated corrective labels are most beneficial in low-data regimes, and (3) label filtering based on estimated dynamics model error enables performance gains. To effectively apply CCIL to robotic domains, we offer a practical instantiation of the framework and insights into design choices and hyperparameter selection. Our work demonstrates CCIL's practicality for alleviating compounding errors in imitation learning on physical robots.

Constructing simulation scenes that are both visually and physically realistic is a problem of practical interest in domains ranging from robotics to computer vision. This problem has become even more relevant as researchers wielding large data-hungry learning methods seek new sources of training data for physical decision-making systems. However, building simulation models is often still done by hand. A graphic designer and a simulation engineer work with predefined assets to construct rich scenes with realistic dynamic and kinematic properties. While this may scale to small numbers of scenes, to achieve the generalization properties that are required for data-driven robotic control, we require a pipeline that is able to synthesize large numbers of realistic scenes, complete with 'natural' kinematic and dynamic structures. To attack this problem, we develop models for inferring structure and generating simulation scenes from natural images, allowing for scalable scene generation from web-scale datasets. To train these image-to-simulation models, we show how controllable text-to-image generative models can be used in generating paired training data that allows for modeling of the inverse problem, mapping from realistic images back to complete scene models. We show how this paradigm allows us to build large datasets of scenes in simulation with semantic and physical realism. We present an integrated end-to-end pipeline that generates simulation scenes complete with articulated kinematic and dynamic structures from real-world images and use these for training robotic control policies. We then robustly deploy in the real world for tasks like articulated object manipulation. In doing so, our work provides both a pipeline for large-scale generation of simulation environments and an integrated system for training robust robotic control policies in the resulting environments., Comment: Accepted at RSS2024

Efficient parameter identification of electrochemical models is crucial for accurate monitoring and control of lithium-ion cells. This process becomes challenging when applied to complex models that rely on a considerable number of interdependent parameters that affect the output response. Gradient-based and metaheuristic optimization techniques, although previously employed for this task, are limited by their lack of robustness, high computational costs, and susceptibility to local minima. In this study, Bayesian Optimization is used for tuning the dynamic parameters of an electrochemical equivalent circuit battery model (E-ECM) for a nickel-manganese-cobalt (NMC)-graphite cell. The performance of the Bayesian Optimization is compared with baseline methods based on gradient-based and metaheuristic approaches. The robustness of the parameter optimization method is tested by performing verification using an experimental drive cycle. The results indicate that Bayesian Optimization outperforms Gradient Descent and PSO optimization techniques, achieving reductions on average testing loss by 28.8% and 5.8%, respectively. Moreover, Bayesian optimization significantly reduces the variance in testing loss by 95.8% and 72.7%, respectively., Comment: 6 pages

In recent years, the field of Transfer Evolutionary Optimization (TrEO) has witnessed substantial growth, fueled by the realization of its profound impact on solving complex problems. Numerous algorithms have emerged to address the challenges posed by transferring knowledge between tasks. However, the recently highlighted ``no free lunch theorem'' in transfer optimization clarifies that no single algorithm reigns supreme across diverse problem types. This paper addresses this conundrum by adopting a benchmarking approach to evaluate the performance of various TrEO algorithms in realistic scenarios. Despite the growing methodological focus on transfer optimization, existing benchmark problems often fall short due to inadequate design, predominantly featuring synthetic problems that lack real-world relevance. This paper pioneers a practical TrEO benchmark suite, integrating problems from the literature categorized based on the three essential aspects of Big Source Task-Instances: volume, variety, and velocity. Our primary objective is to provide a comprehensive analysis of existing TrEO algorithms and pave the way for the development of new approaches to tackle practical challenges. By introducing realistic benchmarks that embody the three dimensions of volume, variety, and velocity, we aim to foster a deeper understanding of algorithmic performance in the face of diverse and complex transfer scenarios. This benchmark suite is poised to serve as a valuable resource for researchers, facilitating the refinement and advancement of TrEO algorithms in the pursuit of solving real-world problems., Comment: 17 pages, 18 figures

Teaching robots novel skills with demonstrations via human-in-the-loop data collection techniques like kinesthetic teaching or teleoperation puts a heavy burden on human supervisors. In contrast to this paradigm, it is often significantly easier to provide raw, action-free visual data of tasks being performed. Moreover, this data can even be mined from video datasets or the web. Ideally, this data can serve to guide robot learning for new tasks in novel environments, informing both "what" to do and "how" to do it. A powerful way to encode both the "what" and the "how" is to infer a well-shaped reward function for reinforcement learning. The challenge is determining how to ground visual demonstration inputs into a well-shaped and informative reward function. We propose a technique Rank2Reward for learning behaviors from videos of tasks being performed without access to any low-level states and actions. We do so by leveraging the videos to learn a reward function that measures incremental "progress" through a task by learning how to temporally rank the video frames in a demonstration. By inferring an appropriate ranking, the reward function is able to guide reinforcement learning by indicating when task progress is being made. This ranking function can be integrated into an adversarial imitation learning scheme resulting in an algorithm that can learn behaviors without exploiting the learned reward function. We demonstrate the effectiveness of Rank2Reward at learning behaviors from raw video on a number of tabletop manipulation tasks in both simulations and on a real-world robotic arm. We also demonstrate how Rank2Reward can be easily extended to be applicable to web-scale video datasets., Comment: ICRA 2024

Model-free control strategies such as reinforcement learning have shown the ability to learn control strategies without requiring an accurate model or simulator of the world. While this is appealing due to the lack of modeling requirements, such methods can be sample inefficient, making them impractical in many real-world domains. On the other hand, model-based control techniques leveraging accurate simulators can circumvent these challenges and use a large amount of cheap simulation data to learn controllers that can effectively transfer to the real world. The challenge with such model-based techniques is the requirement for an extremely accurate simulation, requiring both the specification of appropriate simulation assets and physical parameters. This requires considerable human effort to design for every environment being considered. In this work, we propose a learning system that can leverage a small amount of real-world data to autonomously refine a simulation model and then plan an accurate control strategy that can be deployed in the real world. Our approach critically relies on utilizing an initial (possibly inaccurate) simulator to design effective exploration policies that, when deployed in the real world, collect high-quality data. We demonstrate the efficacy of this paradigm in identifying articulation, mass, and other physical parameters in several challenging robotic manipulation tasks, and illustrate that only a small amount of real-world data can allow for effective sim-to-real transfer. Project website at https://weirdlabuw.github.io/asid, Comment: Project website at https://weirdlabuw.github.io/asid

The objective is to describe a rare case of bilateral Choanal atresia (CA). It is a congenital deformity of posterior nares where both nares are not patent at all or stenosed. While unilateral CA can remain unidentified for years, bilateral CA is typically detectable immediately after birth in neonates who prefer to breathe via their nose until they are four months old. An 18 days old male patient was admitted to the male surgery ward with complaints of respiratory distress and not taking feed for 2 days. The patient has been hospitalized after birth in a railway hospital and referred to a pediatric surgeon and he was diagnosed as a case of bilateral CA. During surgery, posterior nares were perforated, and dilated and stenting of posterior nares was performed with infant feeding tube no.8 with the help of a guide wire. Bilateral CA is a medical emergency that needs to be treated in newborns as early as possible. CT scan of the skull base and paranasal sinus can be helpful as a diagnostic tool in detecting the condition. Keywords: Choanal atresia, stenting, congenital deformity, infant feeding tube

We consider a point process (PP) generated by superimposing an independent Poisson point process (PPP) on each line of a 2D Poisson line process (PLP). Termed PLP-PPP, this PP is suitable for modeling networks formed on an irregular collection of lines, such as vehicles on a network of roads and sensors deployed along trails in a forest. Inspired by vehicular networks in which vehicles connect with their nearest wireless base stations (BSs), we consider a random bipartite associator graph in which each point of the PLP-PPP is associated with the nearest point of an independent PPP through an edge. This graph is equivalent to the partitioning of PLP-PPP by a Poisson Voronoi tessellation (PVT) formed by an independent PPP. We first characterize the exact distribution of the number of points of PLP-PPP falling inside the ball centered at an arbitrary location in as well as the typical point of PLP-PPP. Using these distributions, we derive cumulative distribution functions (CDFs) and probability density functions (PDFs) of kth contact distance (CD) and the nearest neighbor distance (NND) of PLP-PPP. As intermediate results, we present the empirical distribution of the perimeter and approximate distribution of the length of the typical chord of the zero-cell of this PVT. Using these results, we present two close approximations of the distribution of node degree of the random bipartite associator graph. In a vehicular network setting, this result characterizes the number of vehicles connected to each BS, which models its load. Since each BS has to distribute its limited resources across all the vehicles connected to it, a good statistical understanding of load is important for an efficient system design. Several applications of these new results to different wireless network settings are also discus

Els mètodes tradicionals que intenten resoldre tasques a la vida real amb robots mitjançant l'ús de la IA, tenen molts inconvenients. L'aprenentatge per imitació dona bons resultats a l'hora de resoldre tasques molt específiques, però necessita una gran quantitat de demostracions d'experts per poder superar el problema de propagació d'errors. Els mètodes centrats en l'aprenentatge per reforç també presenten un problema fonamental, necessiten una funció de recompensa curosament dissenyada, que pot ser arbitràriament complicada d'especificar, en especial quan tractem amb imatges, i ha de ser densa per a poder resoldre les tasques més complicades. En aquest treball, explorem com els robots poden aprendre en el món real mitjançant aprenentatge autosupervisat a través d'un procés d'anotacions retrospectives. En particular, per aconseguir-ho utilitzem opinions humanes en forma de comparacions binàries, per tal de guiar l'exploració. Aquest tipus d'anotacions es poden recollir asíncronament i de manera molt escalable mitjançant el crowdsourcing. També ensenyem com el nostre mètode és robust als errors en les anotacions i com en necessita molt menys que altres algoritmes. Finalment, ensenyem com podem modificar el nostre mètode perquè pugui ser executat contínuament, sense reinicis. En aquest estudi mostrem com el nostre algoritme supera a altres mètodes clàssics i com pot ser aplicat en el món real., Los métodos tradicionales que tratan de resolver tareas en el mundo real con robots entrenados mediante IA, tienen muchos inconvenientes. El aprendizaje por imitación da buenos resultados a la hora de resolver tareas muy específicas, pero necesita gran cantidad de demostraciones de expertos para poder lidiar con la propagación de errores. Enfoques basados en aprendizaje por refuerzo también presentan un problema fundamental, necesitan una función de recompensa cuidadosamente diseñada, la cual es específica a una tarea en concreto, puede ser arbitrariamente difícil de especificar, especialmente cuando trabajamos con imágenes, y necesita ser densa para permitir resolver las tareas más complicadas. En este trabajo, exploramos como los robots pueden aprender en el mundo real usando aprendizaje autosupervisado a través de un proceso de anotaciones retrospectivas. En particular, para lograrlo usamos opiniones humanas, en forma de comparaciones binarias, para guiar la exploración. Este tipo de anotaciones se pueden recoger asíncronamente y de manera muy escalable usando crowdsourcing. También mostramos como nuestro método es robusto a los errores en las anotaciones y como necesita mucha menos supervisión humana que otros algoritmos. Finalmente, mostramos cómo podemos modificar nuestro algoritmo para que pueda ser ejecutado continuamente, sin reseteos. En este estudio, mostramos como nuestro algoritmo supera a otros métodos clásicos y como puede ser aplicado en la vida real., Traditional methods that aim to solve tasks involving robots in the real world leveraging AI, suffer from many drawbacks: Behaviour Cloning shows great results at solving specific tasks, but it requires a considerably large amount of expert demonstrations to deal with compounding errors. Approaches based on Reinforcement Learning also present a fundamental problem, namely reward shaping, which is task-specific, can be non-trivial to specify, especially when dealing with images, and needs to be dense in order to allow us to solve hard tasks. In this work, we explore how robots can learn in the real world by using self-supervised learning via hindsight relabeling. In particular, to do so, we leverage human feedback, in the form of binary comparisons, to guide exploration. This type of feedback can be collected asynchronously and is easily scalable via crowd-sourcing. We also show that our method is robust to noisy labels and requires less feedback than other algorithms. Finally, we show how we can adapt our algorithm to make it run continuously without resets. In this work, we show how our algorithm outperforms all baselines in some simulation benchmarks and how it can also be used in the real world., Outgoing

The objective is to describe a rare case of bilateral Choanal atresia (CA). It is a congenital deformity of posterior nares where both nares are not patent at all or stenosed. While unilateral CA can remain unidentified for years, bilateral CA is typically detectable immediately after birth in neonates who prefer to breathe via their nose until they are four months old. An 18 days old male patient was admitted to the male surgery ward with complaints of respiratory distress and not taking feed for 2 days. The patient has been hospitalized after birth in a railway hospital and referred to a pediatric surgeon and he was diagnosed as a case of bilateral CA. During surgery, posterior nares were perforated, and dilated and stenting of posterior nares was performed with infant feeding tube no.8 with the help of a guide wire. Bilateral CA is a medical emergency that needs to be treated in newborns as early as possible. CT scan of the skull base and paranasal sinus can be helpful as a diagnostic tool in detecting the condition. Keywords: Choanal atresia, stenting, congenital deformity, infant feeding tube

Soy protein hydrolysates enhance integral viable cell density (IVCD) and recombinant protein production (Immunoglobulin, IgG) in cell cultures, but their functionality varies from batch-to-batch. This is undesirable since it affects both quantity and characteristics of the recombinant proteins. It is hypothesized that the variability of hydrolysates is due to variations in meal and hydrolysate processing treatments. To study this, hydrolysates were produced from meals heated at 121 °C/0–120 min. The heating decreased free amino acid and reducing monosaccharide contents in meals (0.72–0.27% and 3.3–2.6%) and hydrolysates (14.7–7.1% and 16.9–7.9%). Dry heating introduced large variation in the IVCD ((115–316%), but additional heating in suspension reduced it (131–159%). The decrease in IVCD variation corresponded with decreased variation in carboxymethyl-lysine (CML) and lysinoalanine (LAL) contents. Thus, meal and hydrolysate processing induced substantial variation in hydrolysate functionality. It is therefore critical to establish strict process controls for meal and hydrolysate production to ensure consistency.

Wireless communications aided by reconfigurable intelligent surfaces (RISs) is a promising way to improve the coverage for cellular users. The controlled reflection of signals from RISs is especially useful in mm-wave/THz networks when the direct link between a cellular user and its serving base station (BS) is weak or unavailable due to blockages. However, the joint blockage of the user-RIS and the user-BS links may significantly degrade the performance of RIS-aided transmissions. This paper aims to study the impact of joint blockages on downlink performance. When RIS locations are coupled with BS locations, using tools from stochastic geometry, we obtain an optimal placement of RISs to either minimize the joint blockage probability of the user-RIS and the user-BS links or maximize the downlink coverage probability. The results show that installing RISs near the cell edge of BSs usually provides optimal coverage. Moreover, deploying RISs on street intersections improves the coverage probability. For users associated with BSs that are deployed sufficiently close to intersections, the intersection-mounted RISs offer a better coverage performance as compared to BS-coupled RISs.

Grasping small objects surrounded by unstable or non-rigid material plays a crucial role in applications such as surgery, harvesting, construction, disaster recovery, and assisted feeding. This task is especially difficult when fine manipulation is required in the presence of sensor noise and perception errors; errors inevitably trigger dynamic motion, which is challenging to model precisely. Circumventing the difficulty to build accurate models for contacts and dynamics, data-driven methods like reinforcement learning (RL) can optimize task performance via trial and error, reducing the need for accurate models of contacts and dynamics. Applying RL methods to real robots, however, has been hindered by factors such as prohibitively high sample complexity or the high training infrastructure cost for providing resets on hardware. This work presents CherryBot, an RL system that uses chopsticks for fine manipulation that surpasses human reactiveness for some dynamic grasping tasks. By integrating imprecise simulators, suboptimal demonstrations and external state estimation, we study how to make a real-world robot learning system sample efficient and general while reducing the human effort required for supervision. Our system shows continual improvement through 30 minutes of real-world interaction: through reactive retry, it achieves an almost 100% success rate on the demanding task of using chopsticks to grasp small objects swinging in the air. We demonstrate the reactiveness, robustness and generalizability of CherryBot to varying object shapes and dynamics (e.g., external disturbances like wind and human perturbations). Videos are available at https://goodcherrybot.github.io/.

Reinforcement learning algorithms typically struggle in the absence of a dense, well-shaped reward function. Intrinsically motivated exploration methods address this limitation by rewarding agents for visiting novel states or transitions, but these methods offer limited benefits in large environments where most discovered novelty is irrelevant for downstream tasks. We describe a method that uses background knowledge from text corpora to shape exploration. This method, called ELLM (Exploring with LLMs) rewards an agent for achieving goals suggested by a language model prompted with a description of the agent's current state. By leveraging large-scale language model pretraining, ELLM guides agents toward human-meaningful and plausibly useful behaviors without requiring a human in the loop. We evaluate ELLM in the Crafter game environment and the Housekeep robotic simulator, showing that ELLM-trained agents have better coverage of common-sense behaviors during pretraining and usually match or improve performance on a range of downstream tasks. Code available at https://github.com/yuqingd/ellm., Comment: ICML 2023

Robot learning methods have the potential for widespread generalization across tasks, environments, and objects. However, these methods require large diverse datasets that are expensive to collect in real-world robotics settings. For robot learning to generalize, we must be able to leverage sources of data or priors beyond the robot's own experience. In this work, we posit that image-text generative models, which are pre-trained on large corpora of web-scraped data, can serve as such a data source. We show that despite these generative models being trained on largely non-robotics data, they can serve as effective ways to impart priors into the process of robot learning in a way that enables widespread generalization. In particular, we show how pre-trained generative models can serve as effective tools for semantically meaningful data augmentation. By leveraging these pre-trained models for generating appropriate "semantic" data augmentations, we propose a system GenAug that is able to significantly improve policy generalization. We apply GenAug to tabletop manipulation tasks, showing the ability to re-target behavior to novel scenarios, while only requiring marginal amounts of real-world data. We demonstrate the efficacy of this system on a number of object manipulation problems in the real world, showing a 40% improvement in generalization to novel scenes and objects.

Spherical robots can conduct surveillance in hostile, cluttered environments without being damaged, as their protective shell can safely house sensors such as cameras. However, lateral oscillations, also known as wobble, occur when these sphere-shaped robots operate at low speeds, leading to shaky camera feedback. These oscillations in a pendulum-actuated spherical robot are caused by the coupling between the forward and steering motions due to nonholonomic constraints. Designing a controller to limit wobbling in these robots is challenging due to their underactuated nature. We propose a model-based controller to navigate a pendulum-actuated spherical robot using wobble-free turning maneuvers consisting of circular arcs and straight lines. The model is developed using Lagrange-D'Alembert equations and accounts for the coupled forward and steering motions. The model is further analyzed to derive expressions for radius of curvature, precession rate, wobble amplitude, and wobble frequency during circular motions. Finally, we design an input-output feedback linearization-based controller to control the robot's heading direction and wobble. Overall, the proposed controller enables a teleoperator to command a specific forward velocity and pendulum angle as per the desired turning radius while limiting the robot's lateral oscillations to enhance the quality of camera feedback., Comment: The length of the research paper is 20 pages, and it contains 15 graphs or illustrations

A spherical robot has many practical advantages as the entire electronics are protected within a hull and can be carried easily by any Unmanned Aerial Vehicle (UAV). However, its use is limited due to finding mounts for sensors. Pendulum actuated spherical robot provides space for mounting sensors at the yoke. We study the non-linear dynamics of a pendulum-actuated spherical robot to analyze the dynamics of internal assembly (yoke) for mounting sensors. For such robots, we provide a coupled dynamic model that takes care of the relationship between forward and sideways motion. We further demonstrate the effects of wobbling and precession captured by our model when the bot is controlled to execute a turning maneuver while moving with a moderate forward velocity, a practical situation encountered by spherical robots moving in an indoor setting. A simulation setup based on the developed model provides visualization of the spherical robot motion., Comment: The paper has been accepted to the 22nd IFAC International Symposium on Automatic Control in Aerospace (ACA) 2022. It consists of 6 pages and 15 figures

Neil Daksla,1 Ashley Wang,1 Zhaosheng Jin,1 Abhishek Gupta,1 Sergio D Bergese1,2 1Department of Anesthesiology, Stony Brook University Health Science Center, Stony Brook, NY, 11794-8480, USA; 2Department of Neurosurgery, Stony Brook University Health Science Center, Stony Brook, NY, 11794-8480, USACorrespondence: Sergio D Bergese, Department of Anesthesiology, Stony Brook University School of Medicine, Health Sciences Center, Level 4, Room 060, Stony Brook, NY, 11794, USA, Tel +1 631 444-2979, Fax +1 631 444-2907, Email sergio.bergese@stonybrookmedicine.eduAbstract: Despite current advances in acute postoperative pain management, prevalence remains high. Inadequate treatment could lead to poor outcomes and even progression to chronic pain. Opioids have traditionally been the mainstay for treatment of moderate to severe acute pain. However, their use has been associated with opioid-related adverse events (ORAEs), such as respiratory depression, sedation, nausea, vomiting, pruritus, and decreased bowel motility. In addition, their liberal use has been implicated in the current opioid epidemic. As a result, there has been renewed interest in multimodal analgesia to target different mechanisms of action in order to achieve a synergistic effect and minimize opioid usage. Oliceridine is a novel mu-opioid receptor agonist that is part of a new class of biased ligands that selectively activate G-protein signaling and downregulate β-arrestin recruitment. Since G-protein signaling has been associated with analgesia while β-arrestin recruitment has been associated with ORAEs, there is potential for a wider therapeutic window. In this review, we will discuss the clinical evidence behind oliceridine and its potential role in acute postoperative pain management. We have systematically searched the PubMed database using the keywords oliceridine, olinvyk, and trv130. All articles identified were reviewed and evaluated, and all clinical trials were included.Keywords: postoperative

Millimeter-waves' propagation characteristics create prospects for spatial and temporal spectrum sharing in a variety of contexts, including cognitive spectrum sharing (CSS). However, CSS along with omnidirectional sensing, is not efficient at mmWave frequencies due to their directional nature of transmission, as this limits secondary networks' ability to access the spectrum. This inspired us to create an analytical approach using stochastic geometry to examine the implications of directional cognitive sensing in mmWave networks. We explore a scenario where multiple secondary transmitter-receiver pairs coexist with a primary transmitter-receiver pair, forming a cognitive network. The positions of the secondary transmitters are modelled using a homogeneous Poisson point process (PPP) with corresponding secondary receivers located around them. A threshold on directional transmission is imposed on each secondary transmitter in order to limit its interference at the primary receiver. We derive the medium-access-probability of a secondary user along with the fraction of the secondary transmitters active at a time-instant. To understand cognition's feasibility, we derive the coverage probabilities of primary and secondary links. We provide various design insights via numerical results. For example, we investigate the interference-threshold's optimal value while ensuring coverage for both links and its dependence on various parameters. We find that directionality improves both links' performance as a key factor. Further, allowing location-aware secondary directionality can help achieve similar coverage for all secondary links., Comment: 30 pages, 12 figures

Artificial Intelligence (AI) has significantly revolutionized radiology, promising improved patient outcomes and streamlined processes. However, it's critical to ensure the fairness of AI models to prevent stealthy bias and disparities from leading to unequal outcomes. This review discusses the concept of fairness in AI, focusing on bias auditing using the Aequitas toolkit, and its real-world implications in radiology, particularly in disease screening scenarios. Aequitas, an open-source bias audit toolkit, scrutinizes AI models' decisions, identifying hidden biases that may result in disparities across different demographic groups and imaging equipment brands. This toolkit operates on statistical theories, analyzing a large dataset to reveal a model's fairness. It excels in its versatility to handle various variables simultaneously, especially in a field as diverse as radiology. The review explicates essential fairness metrics: Equal and Proportional Parity, False Positive Rate Parity, False Discovery Rate Parity, False Negative Rate Parity, and False Omission Rate Parity. Each metric serves unique purposes and offers different insights. We present hypothetical scenarios to demonstrate their relevance in disease screening settings, and how disparities can lead to significant real-world impacts.

Model-free reinforcement learning algorithms have exhibited great potential in solving single-task sequential decision-making problems with high-dimensional observations and long horizons, but are known to be hard to generalize across tasks. Model-based RL, on the other hand, learns task-agnostic models of the world that naturally enables transfer across different reward functions, but struggles to scale to complex environments due to the compounding error. To get the best of both worlds, we propose a self-supervised reinforcement learning method that enables the transfer of behaviors across tasks with different rewards, while circumventing the challenges of model-based RL. In particular, we show self-supervised pre-training of model-free reinforcement learning with a number of random features as rewards allows implicit modeling of long-horizon environment dynamics. Then, planning techniques like model-predictive control using these implicit models enable fast adaptation to problems with new reward functions. Our method is self-supervised in that it can be trained on offline datasets without reward labels, but can then be quickly deployed on new tasks. We validate that our proposed method enables transfer across tasks on a variety of manipulation and locomotion domains in simulation, opening the door to generalist decision-making agents.

Synthesis of digital artifacts conditioned on user prompts has become an important paradigm facilitating an explosion of use cases with generative AI. However, such models often fail to connect the generated outputs and desired target concepts/preferences implied by the prompts. Current research addressing this limitation has largely focused on enhancing the prompts before output generation or improving the model's performance up front. In contrast, this paper conceptualizes prompt evolution, imparting evolutionary selection pressure and variation during the generative process to produce multiple outputs that satisfy the target concepts/preferences better. We propose a multi-objective instantiation of this broader idea that uses a multi-label image classifier-guided approach. The predicted labels from the classifiers serve as multiple objectives to optimize, with the aim of producing diversified images that meet user preferences. A novelty of our evolutionary algorithm is that the pre-trained generative model gives us implicit mutation operations, leveraging the model's stochastic generative capability to automate the creation of Pareto-optimized images more faithful to user preferences., Comment: To appear in Proceedings of the 2023 IEEE Conference on Artificial Intelligence (CAI'23)

Machine learning systems, especially with overparameterized deep neural networks, can generalize to novel test instances drawn from the same distribution as the training data. However, they fare poorly when evaluated on out-of-support test points. In this work, we tackle the problem of developing machine learning systems that retain the power of overparameterized function approximators while enabling extrapolation to out-of-support test points when possible. This is accomplished by noting that under certain conditions, a "transductive" reparameterization can convert an out-of-support extrapolation problem into a problem of within-support combinatorial generalization. We propose a simple strategy based on bilinear embeddings to enable this type of combinatorial generalization, thereby addressing the out-of-support extrapolation problem under certain conditions. We instantiate a simple, practical algorithm applicable to various supervised learning and imitation learning tasks.

Without the ability to estimate and benchmark AI capability advancements, organizations are left to respond to each change reactively, impeding their ability to build viable mid and long-term strategies. This paper explores the recent growth of forecasting, a political science tool that uses explicit assumptions and quantitative estimation that leads to improved prediction accuracy. Done at the collective level, forecasting can identify and verify talent, enable leaders to build better models of AI advancements and improve inputs into design policy. Successful approaches to forecasting and case studies are examined, revealing a subclass of "superforecasters" who outperform 98% of the population and whose insights will be most reliable. Finally, techniques behind successful forecasting are outlined, including Phillip Tetlock's "Ten Commandments." To adapt to a quickly changing technology landscape, designers and policymakers should consider forecasting as a first line of defense., Comment: 7 pages

AI systems may be better thought of as peers than as tools. This paper explores applications of augmented collective intelligence (ACI) beneficial to collaborative ideation. Design considerations are offered for an experiment that evaluates the performance of hybrid human- AI collectives. The investigation described combines humans and large language models (LLMs) to ideate on increasingly complex topics. A promising real-time collection tool called Polis is examined to facilitate ACI, including case studies from citizen engagement projects in Taiwan and Bowling Green, Kentucky. The authors discuss three challenges to consider when designing an ACI experiment: topic selection, participant selection, and evaluation of results. The paper concludes that researchers should address these challenges to conduct empirical studies of ACI in collaborative ideation., Comment: 5 pages

The polycystic ovary syndrome diagnosis is a problem that can be leveraged using prognostication based learning procedures. Many implementations of PCOS can be seen with Machine Learning but the algorithms have certain limitations in utilizing the processing power graphical processing units. The simple machine learning algorithms can be improved with advanced frameworks using Deep Learning. The Linear Discriminant Analysis is a linear dimensionality reduction algorithm for classification that can be boosted in terms of performance using deep learning with Deep LDA, a transformed version of the traditional LDA. In this result oriented paper we present the Deep LDA implementation with a variation for prognostication of PCOS., Comment: 7 pages, 5 figures. To appear in proceedings of Intelligent Computing and Networking (IC-ICN 2022)

We study the problem of object retrieval in scenarios where visual sensing is absent, object shapes are unknown beforehand and objects can move freely, like grabbing objects out of a drawer. Successful solutions require localizing free objects, identifying specific object instances, and then grasping the identified objects, only using touch feedback. Unlike vision, where cameras can observe the entire scene, touch sensors are local and only observe parts of the scene that are in contact with the manipulator. Moreover, information gathering via touch sensors necessitates applying forces on the touched surface which may disturb the scene itself. Reasoning with touch, therefore, requires careful exploration and integration of information over time -- a challenge we tackle. We present a system capable of using sparse tactile feedback from fingertip touch sensors on a dexterous hand to localize, identify and grasp novel objects without any visual feedback. Videos are available at https://taochenshh.github.io/projects/tactofind., Comment: Accepted in ICRA 2023

In this paper, we scale evolutionary algorithms to high-dimensional optimization problems that deceptively possess a low effective dimensionality (certain dimensions do not significantly affect the objective function). To this end, an instantiation of the multiform optimization paradigm is presented, where multiple low-dimensional counterparts of a target high-dimensional task are generated via random embeddings. Since the exact relationship between the auxiliary (low-dimensional) tasks and the target is a priori unknown, a multiform evolutionary algorithm is developed for unifying all formulations into a single multi-task setting. The resultant joint optimization enables the target task to efficiently reuse solutions evolved across various low-dimensional searches via cross-form genetic transfers, hence speeding up overall convergence characteristics. To validate the overall efficacy of our proposed algorithmic framework, comprehensive experimental studies are carried out on well-known continuous benchmark functions as well as a set of practical problems in the hyper-parameter tuning of machine learning models and deep learning models in classification tasks and Predator-Prey games, respectively., Comment: 12 pages,6 figures

Transfer optimization enables data-efficient optimization of a target task by leveraging experiential priors from related source tasks. This is especially useful in multiobjective optimization settings where a set of trade-off solutions is sought under tight evaluation budgets. In this paper, we introduce a novel concept of inverse transfer in multiobjective optimization. Inverse transfer stands out by employing probabilistic inverse models to map performance vectors in the objective space to population search distributions in task-specific decision space, facilitating knowledge transfer through objective space unification. Building upon this idea, we introduce the first Inverse Transfer Multiobjective Evolutionary Optimizer (invTrEMO). A key highlight of invTrEMO is its ability to harness the common objective functions prevalent in many application areas, even when decision spaces do not precisely align between tasks. This allows invTrEMO to uniquely and effectively utilize information from heterogeneous source tasks as well. Furthermore, invTrEMO yields high-precision inverse models as a significant byproduct, enabling the generation of tailored solutions on-demand based on user preferences. Empirical studies on multi- and many-objective benchmark problems, as well as a practical case study, showcase the faster convergence rate and modelling accuracy of the invTrEMO relative to state-of-the-art evolutionary and Bayesian optimization algorithms. The source code of the invTrEMO is made available at https://github.com/LiuJ-2023/invTrEMO.

This paper considers the change point detection problem under dependent samples. In particular, we provide performance guarantees for the MMD-CUSUM test under exponentially $\alpha$, $\beta$, and fast $\phi$-mixing processes, which significantly expands its utility beyond the i.i.d. and Markovian cases used in previous studies. We obtain lower bounds for average-run-length (ARL) and upper bounds for average-detection-delay (ADD) in terms of the threshold parameter. We show that the MMD-CUSUM test enjoys the same level of performance as the i.i.d. case under fast $\phi$-mixing processes. The MMD-CUSUM test also achieves strong performance under exponentially $\alpha$/$\beta$-mixing processes, which are significantly more relaxed than existing results. The MMD-CUSUM test statistic adapts to different settings without modifications, rendering it a completely data-driven, dependence-agnostic change point detection scheme. Numerical simulations are provided at the end to evaluate our findings., Comment: 20 pages, 4 figures. Accepted by IEEE OJ-CSYS

Physics-informed neural networks (PINNs) are at the forefront of scientific machine learning, making possible the creation of machine intelligence that is cognizant of physical laws and able to accurately simulate them. In this paper, the potential of discovering PINNs that generalize over an entire family of physics tasks is studied, for the first time, through a biological lens of the Baldwin effect. Drawing inspiration from the neurodevelopment of precocial species that have evolved to learn, predict and react quickly to their environment, we envision PINNs that are pre-wired with connection strengths inducing strong biases towards efficient learning of physics. To this end, evolutionary selection pressure (guided by proficiency over a family of tasks) is coupled with lifetime learning (to specialize on a smaller subset of those tasks) to produce PINNs that demonstrate fast and physics-compliant prediction capabilities across a range of empirically challenging problem instances. The Baldwinian approach achieves an order of magnitude improvement in prediction accuracy at a fraction of the computation cost compared to state-of-the-art results with PINNs meta-learned by gradient descent. This paper marks a leap forward in the meta-learning of PINNs as generalizable physics solvers.

We study the problem of modeling a population of agents pursuing unknown goals subject to unknown computational constraints. In standard models of bounded rationality, sub-optimal decision-making is simulated by adding homoscedastic noise to optimal decisions rather than explicitly simulating constrained inference. In this work, we introduce a latent inference budget model (L-IBM) that models agents' computational constraints explicitly, via a latent variable (inferred jointly with a model of agents' goals) that controls the runtime of an iterative inference algorithm. L-IBMs make it possible to learn agent models using data from diverse populations of suboptimal actors. In three modeling tasks -- inferring navigation goals from routes, inferring communicative intents from human utterances, and predicting next moves in human chess games -- we show that L-IBMs match or outperform Boltzmann models of decision-making under uncertainty. Inferred inference budgets are themselves meaningful, efficient to compute, and correlated with measures of player skill, partner skill and task difficulty.

With the increasingly widespread adoption of AI in healthcare, maintaining the accuracy and reliability of AI models in clinical practice has become crucial. In this context, we introduce novel methods for monitoring the performance of radiology AI classification models in practice, addressing the challenges of obtaining real-time ground truth for performance monitoring. We propose two metrics - predictive divergence and temporal stability - to be used for preemptive alerts of AI performance changes. Predictive divergence, measured using Kullback-Leibler and Jensen-Shannon divergences, evaluates model accuracy by comparing predictions with those of two supplementary models. Temporal stability is assessed through a comparison of current predictions against historical moving averages, identifying potential model decay or data drift. This approach was retrospectively validated using chest X-ray data from a single-center imaging clinic, demonstrating its effectiveness in maintaining AI model reliability. By providing continuous, real-time insights into model performance, our system ensures the safe and effective use of AI in clinical decision-making, paving the way for more robust AI integration in healthcare, Comment: 10 pages, 4 figures, 2 tables

Recent decisions by leading AI labs to either open-source their models or to restrict access to their models has sparked debate about whether, and how, increasingly capable AI models should be shared. Open-sourcing in AI typically refers to making model architecture and weights freely and publicly accessible for anyone to modify, study, build on, and use. This offers advantages such as enabling external oversight, accelerating progress, and decentralizing control over AI development and use. However, it also presents a growing potential for misuse and unintended consequences. This paper offers an examination of the risks and benefits of open-sourcing highly capable foundation models. While open-sourcing has historically provided substantial net benefits for most software and AI development processes, we argue that for some highly capable foundation models likely to be developed in the near future, open-sourcing may pose sufficiently extreme risks to outweigh the benefits. In such a case, highly capable foundation models should not be open-sourced, at least not initially. Alternative strategies, including non-open-source model sharing options, are explored. The paper concludes with recommendations for developers, standard-setting bodies, and governments for establishing safe and responsible model sharing practices and preserving open-source benefits where safe., Comment: Official release at https://www.governance.ai/research-paper/open-sourcing-highly-capable-foundation-models

Ideally, we would place a robot in a real-world environment and leave it there improving on its own by gathering more experience autonomously. However, algorithms for autonomous robotic learning have been challenging to realize in the real world. While this has often been attributed to the challenge of sample complexity, even sample-efficient techniques are hampered by two major challenges - the difficulty of providing well "shaped" rewards, and the difficulty of continual reset-free training. In this work, we describe a system for real-world reinforcement learning that enables agents to show continual improvement by training directly in the real world without requiring painstaking effort to hand-design reward functions or reset mechanisms. Our system leverages occasional non-expert human-in-the-loop feedback from remote users to learn informative distance functions to guide exploration while leveraging a simple self-supervised learning algorithm for goal-directed policy learning. We show that in the absence of resets, it is particularly important to account for the current "reachability" of the exploration policy when deciding which regions of the space to explore. Based on this insight, we instantiate a practical learning system - GEAR, which enables robots to simply be placed in real-world environments and left to train autonomously without interruption. The system streams robot experience to a web interface only requiring occasional asynchronous feedback from remote, crowdsourced, non-expert humans in the form of binary comparative feedback. We evaluate this system on a suite of robotic tasks in simulation and demonstrate its effectiveness at learning behaviors both in simulation and the real world. Project website https://guided-exploration-autonomous-rl.github.io/GEAR/., Comment: Project website https://guided-exploration-autonomous-rl.github.io/GEAR

Off-policy dynamic programming (DP) techniques such as $Q$-learning have proven to be important in sequential decision-making problems. In the presence of function approximation, however, these techniques often diverge due to the absence of Bellman completeness in the function classes considered, a crucial condition for the success of DP-based methods. In this paper, we show how off-policy learning techniques based on return-conditioned supervised learning (RCSL) are able to circumvent these challenges of Bellman completeness, converging under significantly more relaxed assumptions inherited from supervised learning. We prove there exists a natural environment in which if one uses two-layer multilayer perceptron as the function approximator, the layer width needs to grow linearly with the state space size to satisfy Bellman completeness while a constant layer width is enough for RCSL. These findings take a step towards explaining the superior empirical performance of RCSL methods compared to DP-based methods in environments with near-optimal datasets. Furthermore, in order to learn from sub-optimal datasets, we propose a simple framework called MBRCSL, granting RCSL methods the ability of dynamic programming to stitch together segments from distinct trajectories. MBRCSL leverages learned dynamics models and forward sampling to accomplish trajectory stitching while avoiding the need for Bellman completeness that plagues all dynamic programming algorithms. We propose both theoretical analysis and experimental evaluation to back these claims, outperforming state-of-the-art model-free and model-based offline RL algorithms across several simulated robotics problems.

We present a new technique to enhance the robustness of imitation learning methods by generating corrective data to account for compounding errors and disturbances. While existing methods rely on interactive expert labeling, additional offline datasets, or domain-specific invariances, our approach requires minimal additional assumptions beyond access to expert data. The key insight is to leverage local continuity in the environment dynamics to generate corrective labels. Our method first constructs a dynamics model from the expert demonstration, encouraging local Lipschitz continuity in the learned model. In locally continuous regions, this model allows us to generate corrective labels within the neighborhood of the demonstrations but beyond the actual set of states and actions in the dataset. Training on this augmented data enhances the agent's ability to recover from perturbations and deal with compounding errors. We demonstrate the effectiveness of our generated labels through experiments in a variety of robotics domains in simulation that have distinct forms of continuity and discontinuity, including classic control problems, drone flying, navigation with high-dimensional sensor observations, legged locomotion, and tabletop manipulation.

Large, high-capacity models trained on diverse datasets have shown remarkable successes on efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led to a consolidation of pretrained models, with general pretrained backbones serving as a starting point for many applications. Can such a consolidation happen in robotics? Conventionally, robotic learning methods train a separate model for every application, every robot, and even every environment. Can we instead train generalist X-robot policy that can be adapted efficiently to new robots, tasks, and environments? In this paper, we provide datasets in standardized data formats and models to make it possible to explore this possibility in the context of robotic manipulation, alongside experimental results that provide an example of effective X-robot policies. We assemble a dataset from 22 different robots collected through a collaboration between 21 institutions, demonstrating 527 skills (160266 tasks). We show that a high-capacity model trained on this data, which we call RT-X, exhibits positive transfer and improves the capabilities of multiple robots by leveraging experience from other platforms. More details can be found on the project website https://robotics-transformer-x.github.io., Comment: Project website: https://robotics-transformer-x.github.io

Real-world robotic tasks stretch over extended horizons and encompass multiple stages. Learning long-horizon manipulation tasks, however, is a long-standing challenge, and demands decomposing the overarching task into several manageable subtasks to facilitate policy learning and generalization to unseen tasks. Prior task decomposition methods require task-specific knowledge, are computationally intensive, and cannot readily be applied to new tasks. To address these shortcomings, we propose Universal Visual Decomposer (UVD), an off-the-shelf task decomposition method for visual long horizon manipulation using pre-trained visual representations designed for robotic control. At a high level, UVD discovers subgoals by detecting phase shifts in the embedding space of the pre-trained representation. Operating purely on visual demonstrations without auxiliary information, UVD can effectively extract visual subgoals embedded in the videos, while incurring zero additional training cost on top of standard visuomotor policy training. Goal-conditioned policies learned with UVD-discovered subgoals exhibit significantly improved compositional generalization at test time to unseen tasks. Furthermore, UVD-discovered subgoals can be used to construct goal-based reward shaping that jump-starts temporally extended exploration for reinforcement learning. We extensively evaluate UVD on both simulation and real-world tasks, and in all cases, UVD substantially outperforms baselines across imitation and reinforcement learning settings on in-domain and out-of-domain task sequences alike, validating the clear advantage of automated visual task decomposition within the simple, compact UVD framework.

We present RoboHive, a comprehensive software platform and ecosystem for research in the field of Robot Learning and Embodied Artificial Intelligence. Our platform encompasses a diverse range of pre-existing and novel environments, including dexterous manipulation with the Shadow Hand, whole-arm manipulation tasks with Franka and Fetch robots, quadruped locomotion, among others. Included environments are organized within and cover multiple domains such as hand manipulation, locomotion, multi-task, multi-agent, muscles, etc. In comparison to prior works, RoboHive offers a streamlined and unified task interface taking dependency on only a minimal set of well-maintained packages, features tasks with high physics fidelity and rich visual diversity, and supports common hardware drivers for real-world deployment. The unified interface of RoboHive offers a convenient and accessible abstraction for algorithmic research in imitation, reinforcement, multi-task, and hierarchical learning. Furthermore, RoboHive includes expert demonstrations and baseline results for most environments, providing a standard for benchmarking and comparisons. Details: https://sites.google.com/view/robohive, Comment: Accepted at 37th Conference on Neural Information Processing Systems (NeurIPS 2023) Track on Datasets and Benchmarks

Offline policy learning is aimed at learning decision-making policies using existing datasets of trajectories without collecting additional data. The primary motivation for using reinforcement learning (RL) instead of supervised learning techniques such as behavior cloning is to find a policy that achieves a higher average return than the trajectories constituting the dataset. However, we empirically find that when a dataset is dominated by suboptimal trajectories, state-of-the-art offline RL algorithms do not substantially improve over the average return of trajectories in the dataset. We argue this is due to an assumption made by current offline RL algorithms of staying close to the trajectories in the dataset. If the dataset primarily consists of sub-optimal trajectories, this assumption forces the policy to mimic the suboptimal actions. We overcome this issue by proposing a sampling strategy that enables the policy to only be constrained to ``good data" rather than all actions in the dataset (i.e., uniform sampling). We present a realization of the sampling strategy and an algorithm that can be used as a plug-and-play module in standard offline RL algorithms. Our evaluation demonstrates significant performance gains in 72 imbalanced datasets, D4RL dataset, and across three different offline RL algorithms. Code is available at https://github.com/Improbable-AI/dw-offline-rl., Comment: Accepted NeurIPS 2023

This paper studies Hoeffding's inequality for Markov chains under the generalized concentrability condition defined via integral probability metric (IPM). The generalized concentrability condition establishes a framework that interpolates and extends the existing hypotheses of Markov chain Hoeffding-type inequalities. The flexibility of our framework allows Hoeffding's inequality to be applied beyond the ergodic Markov chains in the traditional sense. We demonstrate the utility by applying our framework to several non-asymptotic analyses arising from the field of machine learning, including (i) a generalization bound for empirical risk minimization with Markovian samples, (ii) a finite sample guarantee for Ployak-Ruppert averaging of SGD, and (iii) a new regret bound for rested Markovian bandits with general state space.

Large Language Models (LLMs) have been shown to act like planners that can decompose high-level instructions into a sequence of executable instructions. However, current LLM-based planners are only able to operate with a fixed set of skills. We overcome this critical limitation and present a method for using LLM-based planners to query new skills and teach robots these skills in a data and time-efficient manner for rigid object manipulation. Our system can re-use newly acquired skills for future tasks, demonstrating the potential of open world and lifelong learning. We evaluate the proposed framework on multiple tasks in simulation and the real world. Videos are available at: https://sites.google.com/mit.edu/halp-robot-learning.

Dexterous manipulation tasks involving contact-rich interactions pose a significant challenge for both model-based control systems and imitation learning algorithms. The complexity arises from the need for multi-fingered robotic hands to dynamically establish and break contacts, balance non-prehensile forces, and control large degrees of freedom. Reinforcement learning (RL) offers a promising approach due to its general applicability and capacity to autonomously acquire optimal manipulation strategies. However, its real-world application is often hindered by the necessity to generate a large number of samples, reset the environment, and obtain reward signals. In this work, we introduce an efficient system for learning dexterous manipulation skills with RL to alleviate these challenges. The main idea of our approach is the integration of recent advances in sample-efficient RL and replay buffer bootstrapping. This combination allows us to utilize data from different tasks or objects as a starting point for training new tasks, significantly improving learning efficiency. Additionally, our system completes the real-world training cycle by incorporating learned resets via an imitation-based pickup policy as well as learned reward functions, eliminating the need for manual resets and reward engineering. We demonstrate the benefits of reusing past data as replay buffer initialization for new tasks, for instance, the fast acquisition of intricate manipulation skills in the real world on a four-fingered robotic hand. (Videos: https://sites.google.com/view/reboot-dexterous), Comment: Accepted at CORL 2023. The first two authors contributed equally