Search

Influenced by ChatGPT, artificial intelligence (AI) large models have witnessed a global upsurge in large model research and development. As people enjoy the convenience by this AI large model, more and more large models in subdivided fields are gradually being proposed, especially large models in radiology imaging field. This article first introduces the development history of large models, technical details, workflow, working principles of multimodal large models and working principles of video generation large models. Secondly, we summarize the latest research progress of AI large models in radiology education, radiology report generation, applications of unimodal and multimodal radiology. Finally, this paper also summarizes some of the challenges of large AI models in radiology, with the aim of better promoting the rapid revolution in the field of radiography.

In the domain of causal inference research, the prevalent potential outcomes framework, notably the Rubin Causal Model (RCM), often overlooks individual interference and assumes independent treatment effects. This assumption, however, is frequently misaligned with the intricate realities of real-world scenarios, where interference is not merely a possibility but a common occurrence. Our research endeavors to address this discrepancy by focusing on the estimation of direct and spillover treatment effects under two assumptions: (1) network-based interference, where treatments on neighbors within connected networks affect one's outcomes, and (2) non-random treatment assignments influenced by confounders. To improve the efficiency of estimating potentially complex effects functions, we introduce an novel active learning approach: Active Learning in Causal Inference with Interference (ACI). This approach uses Gaussian process to flexibly model the direct and spillover treatment effects as a function of a continuous measure of neighbors' treatment assignment. The ACI framework sequentially identifies the experimental settings that demand further data. It further optimizes the treatment assignments under the network interference structure using genetic algorithms to achieve efficient learning outcome. By applying our method to simulation data and a Tencent game dataset, we demonstrate its feasibility in achieving accurate effects estimations with reduced data requirements. This ACI approach marks a significant advancement in the realm of data efficiency for causal inference, offering a robust and efficient alternative to traditional methodologies, particularly in scenarios characterized by complex interference patterns., Comment: conference paper

The evolving landscape of online multiplayer gaming presents unique challenges in assessing the causal impacts of game features. Traditional A/B testing methodologies fall short due to complex player interactions, leading to violations of fundamental assumptions like the Stable Unit Treatment Value Assumption (SUTVA). Unlike traditional social networks with stable and long-term connections, networks in online games are often dynamic and short-lived. Players are temporarily teamed up for the duration of a game, forming transient networks that dissolve once the game ends. This fleeting nature of interactions presents a new challenge compared with running experiments in a stable social network. This study introduces a novel framework for treatment effect estimation in online gaming environments, considering the dynamic and ephemeral network interference that occurs among players. We propose an innovative estimator tailored for scenarios where a completely randomized experimental design is implemented without explicit knowledge of network structures. Notably, our method facilitates post-hoc interference adjustment on experimental data, significantly reducing the complexities and costs associated with intricate experimental designs and randomization strategies. The proposed framework stands out for its ability to accommodate varying levels of interference, thereby yielding more accurate and robust estimations. Through comprehensive simulations set against a variety of interference scenarios, along with empirical validation using real-world data from a mobile gaming environment, we demonstrate the efficacy of our approach. This study represents a pioneering effort in exploring causal inference in user-randomized experiments impacted by dynamic network effects.

Due to the hydrolytic instability of $LiPF_6$ in carbonate-based solvents, HF is a typical impurity in Li-ion battery electrolytes. HF significantly influences the performance of Li-ion batteries, for example by impacting the formation of the solid electrolyte interphase at the anode and by affecting transition metal dissolution at the cathode. Additionally, HF complicates studying fundamental interfacial electrochemistry of Li-ion battery electrolytes, such as direct anion reduction, because it is electrocatalytically relatively unstable, resulting in LiF passivation layers. Methods to selectively remove ppm levels of HF from $LiPF_6$-containing carbonate-based electrolytes are limited. We introduce and benchmark a simple yet efficient electrochemical in situ method to selectively remove ppm amounts of HF from $LiPF_6$-containing carbonate-based electrolytes. The basic idea is the application of a suitable potential to a high surface-area metallic electrode upon which only HF reacts (electrocatalytically) while all other electrolyte components are unaffected under the respective conditions., Comment: 31 pages, 19 figures

Fractional calculus with respect to function $\psi$, also named as $\psi$-fractional calculus, generalizes the Hadamard and the Riemann-Liouville fractional calculi, which causes challenge in numerical treatment. In this paper we study spectral-type methods using mapped Jacobi functions (MJFs) as basis functions and obtain efficient algorithms to solve $\psi$-fractional differential equations. In particular, we setup the Petrov-Galerkin spectral method and spectral collocation method for initial and boundary value problems involving $\psi$-fractional derivatives. We develop basic approximation theory for the MJFs and conduct the error estimates of the derived methods. We also establish a recurrence relation to evaluate the collocation differentiation matrix for implementing the spectral collocation algorithm. Numerical examples confirm the theoretical results and demonstrate the effectiveness of the spectral and collocation methods., Comment: This is a full length version of a submission to TWMS

Diffusion models have become a mainstream approach for high-resolution image synthesis. However, directly generating higher-resolution images from pretrained diffusion models will encounter unreasonable object duplication and exponentially increase the generation time. In this paper, we discover that object duplication arises from feature duplication in the deep blocks of the U-Net. Concurrently, We pinpoint the extended generation times to self-attention redundancy in U-Net's top blocks. To address these issues, we propose a tuning-free higher-resolution framework named HiDiffusion. Specifically, HiDiffusion contains Resolution-Aware U-Net (RAU-Net) that dynamically adjusts the feature map size to resolve object duplication and engages Modified Shifted Window Multi-head Self-Attention (MSW-MSA) that utilizes optimized window attention to reduce computations. we can integrate HiDiffusion into various pretrained diffusion models to scale image generation resolutions even to 4096x4096 at 1.5-6x the inference speed of previous methods. Extensive experiments demonstrate that our approach can address object duplication and heavy computation issues, achieving state-of-the-art performance on higher-resolution image synthesis tasks.

Miniaturized spectrometry system is playing an essential role for materials analysis in the development of in-situ or portable sensing platforms across research and industry. However, there unavoidably exists trade-offs between the resolution and operation bandwidth as the device scale down. Here, we report an extreme miniaturized computational photonic molecule (PM) spectrometer utilizing the diverse spectral characteristics and mode-hybridization effect of split eigenfrequencies and super-modes, which effectively eliminates the inherent periodicity and expands operation bandwidth with ultra-high spectral resolution. These results of dynamic control of the frequency, amplitude, and phase of photons in the photonic multi-atomic systems, pave the way to the development of benchtop sensing platforms for applications previously unfeasible due to resolution-bandwidth-footprint limitations, such as in gas sensing or nanoscale biomedical sensing.

On-chip implementation of optical nonlinear activation functions (NAFs) is essential for realizing large-scale photonic neural chips. To implement different neural processing and machine learning tasks with optimal performances, different NAFs are explored with the use of different devices. From the perspective of on-chip integration and reconfigurability of photonic neural network (PNN), it is highly preferred that a single compact device can fulfill multiple NAFs. Here, we propose and experimentally demonstrate a compact high-speed microdisk modulator to realize multiple NAFs. The fabricated microdisk modulator has an add-drop configuration in which a lateral PN junction is incorporated for tuning. Based on high-speed nonlinear electrical-optical (E-O) effect, multiple NAFs are realized by electrically controlling free-carrier injection. Thanks to its strong optical confinement of the disk cavity, all-optical thermo-optic (TO) nonlinear effect can also be leveraged to realize other four different NAFs, which is difficult to be realized with the use of electrical-optical effect. With the use of the realized nonlinear activation function, a convolutional neural network (CNN) is studied to perform handwritten digit classification task, and an accuracy as large as 98% is demonstrated, which verifies the effectiveness of the use of the high-speed microdisk modulator to realize the NAFs. Thanks to its compact footprint and strong electrical-optical or all-optical effects, the microdisk modulator features multiple NAFs, which could serve as a flexible nonlinear unit for large-scale PNNs.

We present a study of the measurement of the effective weak mixing angle parameter ($\sin^2\theta^{\ell}_\text{eff}$) at the Circular Electron Positron Collider (CEPC). As a fundamental physics parameter, $\sin^2\theta^{\ell}_\text{eff}$ plays a key role not only in the global test of the standard model electroweak sector, but also in constraining the potential beyond standard model new physics at high energy frontier. CEPC proposes a two year running period around the Z boson mass pole at high instataneous luminosity, providing a large data sample with $4\times 10^{12}$ $Z$ candidates generated in total. It allows a high precision measurement of $\sin^2\theta^{\ell}_\text{eff}$ both in the lepton and quark final states, of which the uncertainty can be one order of magnitude lower than any previous measurement at the LEP, SLC, Tevatron and LHC. It will not only improve the overall precision of the $\sin^2\theta^\ell_\text{eff}$ experimental determination to be comparable to the preicision of the theoretical calculation with two-loop radiative corrections, but also provide direct comparisons between different final states. In this paper, we also study the measurement of $\sin^2\theta^{\ell}_\text{eff}$ at high mass region. With one month data taken, the precision of $\sin^2\theta^{\ell}_\text{eff}$ measured at 130 GeV from $b$ quark final state is 0.00010, which will be an important experimental observation on the energy-running effect of $\sin^2\theta^{\ell}_\text{eff}$.

In-circuit impedance measurement provides useful information for many EMC applications. The inductive coupling approach is a promising in-circuit impedance measurement method due to its non-contact characteristics and simple on-site implementation. Many measurement setups of this approach were reported. However, a comprehensive survey and comparison of these setups have not been found in the literature. This paper reviews these setups in terms of their characteristics, limitations, and applications. In addition, recommendations for future research are also presented., Comment: The paper has been accepted to be presented at the 2022 Asia-Pacific International Symposium on Electromagnetic Compatibility (APEMC) in July 2022

Impedance is one of the vital parameters that provides useful information for many power electronics related applications. A lot of impedance measurement methods in power electronics have been reported. However, a comprehensive investigation among these methods in terms of their characteristics, advantages, and limitations has not been found in the literature. In order to bridge this gap, a survey of the impedance measurement methods is conducted in this paper. These methods are introduced, discussed, and then classified into different categories depending on the measurement modes, principles, and instruments. Moreover, recommendations for the future research on the impedance measurement are also presented., Comment: The paper has been accepted to be presented at the 2022 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), and will be published in the IEEE Xplore later

In-circuit impedance provides key information for many EMC applications. The inductive coupling approach is a promising method for in-circuit impedance measurement because its measurement setups have no direct electrical contact with the energized system under test, thus greatly simplifying the on-site implementation. This paper presents and summaries the latest research on the inductive coupling approach and its EMC applications. First of all, three common measurement setups for this approach and their respective pros and cons are discussed. Subsequently, their EMC applications are introduced. Finally, recommendations for future research are listed.

Online controlled experimentation is widely adopted for evaluating new features in the rapid development cycle for web products and mobile applications. Measurement of the overall experiment sample is a common practice to quantify the overall treatment effect. In order to understand why the treatment effect occurs in a certain way, segmentation becomes a valuable approach to a finer analysis of experiment results. This paper introduces a framework for creating and utilizing user behavioral segments in online experimentation. By using the data of user engagement with individual product components as input, this method defines segments that are closely related to the features being evaluated in the product development cycle. With a real-world example, we demonstrate that the analysis with such behavioral segments offered deep, actionable insights that successfully informed product decision-making., Comment: 2017 IEEE International Conference on Data Science and Advanced Analytics (DSAA). IEEE, 2017

Smart home IoT systems and devices are susceptible to attacks and malfunctions. As a result, users' concerns about their security and safety issues arise along with the prevalence of smart home deployments. In a smart home, various anomalies (such as fire or flooding) could happen, due to cyber attacks, device malfunctions, or human mistakes. These concerns motivate researchers to propose various anomaly detection approaches. Existing works on smart home anomaly detection focus on checking the sequence of IoT devices' events but leave out the temporal information of events. This limitation prevents them to detect anomalies that cause delay rather than missing/injecting events. To fill this gap, in this paper, we propose a novel anomaly detection method that takes the inter-event intervals into consideration. We propose an innovative metric to quantify the temporal similarity between two event sequences. We design a mechanism to learn the temporal patterns of event sequences of common daily activities. Delay-caused anomalies are detected by comparing the sequence with the learned patterns. We collect device events from a real-world testbed for training and testing. The experiment results show that our proposed method achieves accuracies of 93%, 88%, 89% for three daily activities.

This paper investigates the influence of different motor stator failures on the common-mode (CM) current. The stator winding failures will affect the motor's CM impedance, and also increase the unbalance of the differential-mode (DM) noise path. The experimental results show that the former will increase the CM current induced by the CM noise source, while the latter will cause the DM current induced by the DM noise source to be converted into the CM current., Comment: The conference paper has been accepted by APEMC 2021, which is to be published in IEEE Xplore soon

The in-circuit common-mode (CM) impedance at the AC input of a motor drive system (MDS) provides valuable inputs for evaluating and estimating the CM electromagnetic interference (EMI) noise generated by the switching of power semiconductor devices in the MDS. This paper introduces a single-probe setup (SPS) with frequency-domain measurement to extract the in-circuit CM impedance of a MDS under its different operating modes. The SPS has the merits of non-contact measurement and simple structure., Comment: This is a modified/final version of arXiv:2110.04068

The in-circuit differential-mode (DM) impedance at the AC input of a motor drive system (MDS) serves as a key parameter to evaluate and estimate the DM electromagnetic interference (EMI) noise caused by the switching of power semiconductor devices in the MDS. This paper discusses a single-probe setup (SPS) with frequency-domain measurement to extract the in-circuit DM impedance of an MDS under its different operating modes. The advantages of the SPS are its non-contact measurement and simple structure.

This thesis is organized as follows: Chapter 1 introduces the background, motivation, objectives, and contributions of this thesis. Chapter 2 presents a review of existing online impedance extraction approaches. Chapter 3 proposes the improved measurement setup of the inductive coupling approach and introduces the theory behind time-variant online impedance extraction. Chapter 4 develops a three-term calibration technique for the proposed measurement setup to deembed the effect of the probe-to-probe coupling between the inductive probes with the objective to improve the accuracy of online impedance extraction. Chapter 5 discusses the additional measurement setup consideration in industrial applications where significant electrical noise and power surges are present. Chapter 6 discusses and demonstrates the application of the inductive coupling approach in online detection of the incipient stator faults in the inverter-fed induction motor. Chapter 7 further extends the application of this approach for non-intrusive extraction of the voltage-dependent capacitances of the silicon carbide (SiC) power metal-oxide-semiconductor field-effect transistor (MOSFET). Finally, Chapter 8 concludes this thesis and proposes future works that are worth exploring.

Uplift modeling is a causal learning technique that estimates subgroup-level treatment effects. It is commonly used in industry and elsewhere for tasks such as targeting ads. In a typical setting, uplift models can take thousands of features as inputs, which is costly and results in problems such as overfitting and poor model interpretability. Consequently, there is a need to select a subset of the most important features for modeling. However, traditional methods for doing feature selection are not fit for the task because they are designed for standard machine learning models whose target is importantly different from uplift models. To address this, we introduce a set of feature selection methods explicitly designed for uplift modeling, drawing inspiration from statistics and information theory. We conduct empirical evaluations on the proposed methods on publicly available datasets, demonstrating the advantages of the proposed methods compared to traditional feature selection. We make the proposed methods publicly available as a part of the CausalML open-source package.

CausalML is a Python implementation of algorithms related to causal inference and machine learning. Algorithms combining causal inference and machine learning have been a trending topic in recent years. This package tries to bridge the gap between theoretical work on methodology and practical applications by making a collection of methods in this field available in Python. This paper introduces the key concepts, scope, and use cases of this package.

In machine learning applications for online product offerings and marketing strategies, there are often hundreds or thousands of features available to build such models. Feature selection is one essential method in such applications for multiple objectives: improving the prediction accuracy by eliminating irrelevant features, accelerating the model training and prediction speed, reducing the monitoring and maintenance workload for feature data pipeline, and providing better model interpretation and diagnosis capability. However, selecting an optimal feature subset from a large feature space is considered as an NP-complete problem. The mRMR (Minimum Redundancy and Maximum Relevance) feature selection framework solves this problem by selecting the relevant features while controlling for the redundancy within the selected features. This paper describes the approach to extend, evaluate, and implement the mRMR feature selection methods for classification problem in a marketing machine learning platform at Uber that automates creation and deployment of targeting and personalization models at scale. This study first extends the existing mRMR methods by introducing a non-linear feature redundancy measure and a model-based feature relevance measure. Then an extensive empirical evaluation is performed for eight different feature selection methods, using one synthetic dataset and three real-world marketing datasets at Uber to cover different use cases. Based on the empirical results, the selected mRMR method is implemented in production for the marketing machine learning platform. A description of the production implementation is provided and an online experiment deployed through the platform is discussed.

Uplift modeling is an emerging machine learning approach for estimating the treatment effect at an individual or subgroup level. It can be used for optimizing the performance of interventions such as marketing campaigns and product designs. Uplift modeling can be used to estimate which users are likely to benefit from a treatment and then prioritize delivering or promoting the preferred experience to those users. An important but so far neglected use case for uplift modeling is an experiment with multiple treatment groups that have different costs, such as for example when different communication channels and promotion types are tested simultaneously. In this paper, we extend standard uplift models to support multiple treatment groups with different costs. We evaluate the performance of the proposed models using both synthetic and real data. We also describe a production implementation of the approach.

During the rapid development cycle for Internet products (websites and mobile apps), new features are developed and rolled out to users constantly. Features with code defects or design flaws can cause outages and significant degradation of user experience. The traditional method of code review and change management can be time-consuming and error-prone. In order to make the feature rollout process safe and fast, this paper proposes a methodology for rolling out features in an automated way using an adaptive experimental design. Under this framework, a feature is gradually ramped up from a small proportion of users to a larger population based on real-time evaluation of the performance of important metrics. If there are any regression detected during the ramp-up step, the ramp-up process stops and the feature developer is alerted. There are two main algorithm components powering this framework: 1) a continuous monitoring algorithm - using a variant of the sequential probability ratio test (SPRT) to monitor the feature performance metrics and alert feature developers when a metric degradation is detected, 2) an automated ramp-up algorithm - deciding when and how to ramp up to the next stage with larger sample size. This paper presents one monitoring algorithm and three ramping up algorithms including time-based, power-based, and risk-based (a Bayesian approach) schedules. These algorithms are evaluated and compared on both simulated data and real data. There are three benefits provided by this framework for feature rollout: 1) for defective features, it can detect the regression early and reduce negative effect, 2) for healthy features, it rolls out the feature quickly, 3) it reduces the need for manual intervention via the automation of the feature rollout process.

We propose a new majorization-minimization (MM) method for non-smooth and non-convex programs, which is general enough to include the existing MM methods. Besides the local majorization condition, we only require that the difference between the directional derivatives of the objective function and its surrogate function vanishes when the number of iterations approaches infinity, which is a very weak condition. So our method can use a surrogate function that directly approximates the non-smooth objective function. In comparison, all the existing MM methods construct the surrogate function by approximating the smooth component of the objective function. We apply our relaxed MM methods to the robust matrix factorization (RMF) problem with different regularizations, where our locally majorant algorithm shows advantages over the state-of-the-art approaches for RMF. This is the first algorithm for RMF ensuring, without extra assumptions, that any limit point of the iterates is a stationary point., Comment: AAAI16