Search

Federated learning (FL) enables a loose set of participating clients to collaboratively learn a global model via coordination by a central server and with no need for data sharing. Existing FL approaches that rely on complex algorithms with massive models, such as deep neural networks (DNNs), suffer from computation and communication bottlenecks. In this paper, we first propose FedHDC, a federated learning framework based on hyperdimensional computing (HDC). FedHDC allows for fast and light-weight local training on clients, provides robust learning, and has smaller model communication overhead compared to learning with DNNs. However, current HDC algorithms get poor accuracy when classifying larger & more complex images, such as CIFAR10. To address this issue, we design FHDnn, which complements FedHDC with a self-supervised contrastive learning feature extractor. We avoid the transmission of the DNN and instead train only the HDC learner in a federated manner, which accelerates learning, reduces transmission cost, and utilizes the robustness of HDC to tackle network errors. We present a formal analysis of the algorithm and derive its convergence rate both theoretically, and show experimentally that FHDnn converges 3$\times$ faster vs. DNNs. The strategies we propose to improve the communication efficiency enable our design to reduce communication costs by 66$\times$ vs. DNNs, local client compute and energy consumption by ~1.5 - 6$\times$, while being highly robust to network errors. Finally, our proposed strategies for improving the communication efficiency have up to 32$\times$ lower communication costs with good accuracy., Comment: Submitted for publication, 20 pages

Advances in high-throughput microscopy have enabled the rapid acquisition of large numbers of high-content microscopy images. Whether by deep learning or classical algorithms, image analysis pipelines then produce single-cell features. To process these single-cells for downstream applications, we present Pycytominer, a user-friendly, open-source python package that implements the bioinformatics steps, known as image-based profiling. We demonstrate Pycytominers usefulness in a machine learning project to predict nuisance compounds that cause undesirable cell injuries., Comment: We updated: Figures (e.g., remove panel from Figure 1) to increase clarity. Consolidated the introduction, results, and discussion into a single section. Added a new analysis to predict compounds that cause undesirable cell injuries. Added three tables including one to highlight image-based profiling software limitations. 14 pages, 2 main figures, 5 supplementary figures, 3 tables

Large Language Models (LLMs) excel at tackling various natural language tasks. However, due to the significant costs involved in re-training or fine-tuning them, they remain largely static and difficult to personalize. Nevertheless, a variety of applications could benefit from generations that are tailored to users' preferences, goals, and knowledge. Among them is web search, where knowing what a user is trying to accomplish, what they care about, and what they know can lead to improved search experiences. In this work, we propose a novel and general approach that augments an LLM with relevant context from users' interaction histories with a search engine in order to personalize its outputs. Specifically, we construct an entity-centric knowledge store for each user based on their search and browsing activities on the web, which is then leveraged to provide contextually relevant LLM prompt augmentations. This knowledge store is light-weight, since it only produces user-specific aggregate projections of interests and knowledge onto public knowledge graphs, and leverages existing search log infrastructure, thereby mitigating the privacy, compliance, and scalability concerns associated with building deep user profiles for personalization. We validate our approach on the task of contextual query suggestion, which requires understanding not only the user's current search context but also what they historically know and care about. Through a number of experiments based on human evaluation, we show that our approach is significantly better than several other LLM-powered baselines, generating query suggestions that are contextually more relevant, personalized, and useful., Comment: The Web Conference (WWW) 2024

The Proton Improvement Plan II (PIP-II) project is an essential upgrade to Fermilab's particle accelerator complex to enable the world's most intense neutrino beam for LBNF/DUNE and a broad particle physics program for many decades to come. PIP-II will deliver 1.2 MW of proton beam power from the Main Injector, upgradeable to multi-MW capability. The central element of PIP-II is an 800 MeV superconducting radio frequency (SRF) linac, which comprises a room temperature front end followed by an SRF section. The SRF section consists of five different flavors of cavities/cryomodules, including Half Wave Resonators (HWR), Single Spoke and elliptical resonators operating at, or above, state-of-the-art parameters. The first two PIP-II cryomodules, Half Wave Resonator (HWR) and Single Spoke Resonator 1 (SSR1) were installed in the PIP-II Injector Test facility (PIP2IT) and have accelerated beam to above 17 MeV. PIP-II is the first U.S. accelerator project that will be constructed with significant contributions from international partners, including India, Italy, France, United Kingdom and Poland. The project was baselined in April 2022, and the construction phase is underway., Comment: 21st Intl Conf Radio Frequency Superconductivity (SRF 2023)

The recent emergence of NVMe flash devices with Zoned Namespace support, ZNS SSDs, represents a significant new advancement in flash storage. ZNS SSDs introduce a new storage abstraction of append-only zones with a set of new I/O (i.e., append) and management (zone state machine transition) commands. With the new abstraction and commands, ZNS SSDs offer more control to the host software stack than a non-zoned SSD for flash management, which is known to be complex (because of garbage collection, scheduling, block allocation, parallelism management, overprovisioning). ZNS SSDs are, consequently, gaining adoption in a variety of applications (e.g., file systems, key-value stores, and databases), particularly latency-sensitive big-data applications. Despite this enthusiasm, there has yet to be a systematic characterization of ZNS SSD performance with its zoned storage model abstractions and I/O operations. This work addresses this crucial shortcoming. We report on the performance features of a commercially available ZNS SSD (13 key observations), explain how these features can be incorporated into publicly available state-of-the-art ZNS emulators, and recommend guidelines for ZNS SSD application developers. All artifacts (code and data sets) of this study are publicly available at https://github.com/stonet-research/NVMeBenchmarks., Comment: Paper to appear in the https://clustercomp.org/2023/program/

Positioned between pre-training and user deployment, aligning large language models (LLMs) through reinforcement learning (RL) has emerged as a prevailing strategy for training instruction following-models such as ChatGPT. In this work, we initiate the study of privacy-preserving alignment of LLMs through Differential Privacy (DP) in conjunction with RL. Following the influential work of Ziegler et al. (2020), we study two dominant paradigms: (i) alignment via RL without human in the loop (e.g., positive review generation) and (ii) alignment via RL from human feedback (RLHF) (e.g., summarization in a human-preferred way). We give a new DP framework to achieve alignment via RL, and prove its correctness. Our experimental results validate the effectiveness of our approach, offering competitive utility while ensuring strong privacy protections., Comment: Accepted at ICLR 2024

We study the ability of state-of-the art models to answer constraint satisfaction queries for information retrieval (e.g., 'a list of ice cream shops in San Diego'). In the past, such queries were considered to be tasks that could only be solved via web-search or knowledge bases. More recently, large language models (LLMs) have demonstrated initial emergent abilities in this task. However, many current retrieval benchmarks are either saturated or do not measure constraint satisfaction. Motivated by rising concerns around factual incorrectness and hallucinations of LLMs, we present KITAB, a new dataset for measuring constraint satisfaction abilities of language models. KITAB consists of book-related data across more than 600 authors and 13,000 queries, and also offers an associated dynamic data collection and constraint verification approach for acquiring similar test data for other authors. Our extended experiments on GPT4 and GPT3.5 characterize and decouple common failure modes across dimensions such as information popularity, constraint types, and context availability. Results show that in the absence of context, models exhibit severe limitations as measured by irrelevant information, factual errors, and incompleteness, many of which exacerbate as information popularity decreases. While context availability mitigates irrelevant information, it is not helpful for satisfying constraints, identifying fundamental barriers to constraint satisfaction. We open source our contributions to foster further research on improving constraint satisfaction abilities of future models., Comment: 23 pages

Modeling groundwater levels continuously across California's Central Valley (CV) hydrological system is challenging due to low-quality well data which is sparsely and noisily sampled across time and space. The lack of consistent well data makes it difficult to evaluate the impact of 2017 and 2019 wet years on CV groundwater following a severe drought during 2012-2015. A novel machine learning method is formulated for modeling groundwater levels by learning from a 3D lithological texture model of the CV aquifer. The proposed formulation performs multivariate regression by combining Gaussian processes (GP) and deep neural networks (DNN). The hierarchical modeling approach constitutes training the DNN to learn a lithologically informed latent space where non-parametric regression with GP is performed. We demonstrate the efficacy of GP-DNN regression for modeling non-stationary features in the well data with fast and reliable uncertainty quantification, as validated to be statistically consistent with the empirical data distribution from 90 blind wells across CV. We show how the model predictions may be used to supplement hydrological understanding of aquifer responses in basins with irregular well data. Our results indicate that on average the 2017 and 2019 wet years in California were largely ineffective in replenishing the groundwater loss caused during previous drought years.

The properties of correlated electron materials are often intricately linked to Van Hove singularities (VHs) in the vicinity of the Fermi energy. The class of these VHs is of great importance, with higher order ones -- with power-law divergence in the density of states -- leaving frequently distinct signatures in physical properties. We use a new theoretical method to detect and analyse higher order Van Hove singularities (HOVHs) in two-dimensional materials and apply it to the electronic structure of the surface layer of Sr$_2$RuO$_4$. We then constrain a low energy model of the VHs of the surface layer of Sr$_2$RuO$_4$ against angle-resolved photoemission spectroscopy and quasiparticle interference data to analyse the VHs near the Fermi level. We show how these VHs can be engineered into HOVHs., Comment: 8 pages including Supplemental Material, 5 figures

Membership Inference Attacks (MIAs) aim to identify specific data samples within the private training dataset of machine learning models, leading to serious privacy violations and other sophisticated threats. Many practical black-box MIAs require query access to the data distribution (the same distribution where the private data is drawn) to train shadow models. By doing so, the adversary obtains models trained "with" or "without" samples drawn from the distribution, and analyzes the characteristics of the samples under consideration. The adversary is often required to train more than hundreds of shadow models to extract the signals needed for MIAs; this becomes the computational overhead of MIAs. In this paper, we propose that by strategically choosing the samples, MI adversaries can maximize their attack success while minimizing the number of shadow models. First, our motivational experiments suggest memorization as the key property explaining disparate sample vulnerability to MIAs. We formalize this through a theoretical bound that connects MI advantage with memorization. Second, we show sample complexity bounds that connect the number of shadow models needed for MIAs with memorization. Lastly, we confirm our theoretical arguments with comprehensive experiments; by utilizing samples with high memorization scores, the adversary can (a) significantly improve its efficacy regardless of the MIA used, and (b) reduce the number of shadow models by nearly two orders of magnitude compared to state-of-the-art approaches.

Machine learning (ML) - based software systems are rapidly gaining adoption across various domains, making it increasingly essential to ensure they perform as intended. This report presents best practices for the Test and Evaluation (T&E) of ML-enabled software systems across its lifecycle. We categorize the lifecycle of ML-enabled software systems into three stages: component, integration and deployment, and post-deployment. At the component level, the primary objective is to test and evaluate the ML model as a standalone component. Next, in the integration and deployment stage, the goal is to evaluate an integrated ML-enabled system consisting of both ML and non-ML components. Finally, once the ML-enabled software system is deployed and operationalized, the T&E objective is to ensure the system performs as intended. Maintenance activities for ML-enabled software systems span the lifecycle and involve maintaining various assets of ML-enabled software systems. Given its unique characteristics, the T&E of ML-enabled software systems is challenging. While significant research has been reported on T&E at the component level, limited work is reported on T&E in the remaining two stages. Furthermore, in many cases, there is a lack of systematic T&E strategies throughout the ML-enabled system's lifecycle. This leads practitioners to resort to ad-hoc T&E practices, which can undermine user confidence in the reliability of ML-enabled software systems. New systematic testing approaches, adequacy measurements, and metrics are required to address the T&E challenges across all stages of the ML-enabled system lifecycle.

Large language models (LLMs) are documented to struggle in settings that require complex reasoning. Nevertheless, instructing the model to break down the problem into smaller reasoning steps, or ensembling various generations through modifying decoding steps boosts performance. However, these methods assume that the input prompt is fixed and expect the decoding strategies to introduce the diversity needed for ensembling. In this work, we discuss how one can create and leverage variations of the input prompt as a means of diversity of thought. We propose a method that automatically improves prompt diversity by soliciting feedback from the LLM to ideate approaches that are apt for the problem. We then ensemble the diverse prompts in our method DIVSE (DIVerse reasoning path Self-Ensemble) across multiple inference calls, or use diverse approaches within a single inference call; we call the latter IDIV-SE (In-call DIVerse reasoning path Self-Ensemble). Apart from our approaches outperforming prior work, DIV-SE(in particular) advances state-of-the-art performance on the challenging planning and graph coloring benchmarks. Our results improve the Pareto frontier of the accuracy-cost trade-off.

Large language models (LLMs) frequently hallucinate on abstractive summarization tasks such as document-based question-answering, meeting summarization, and clinical report generation, even though all necessary information is included in context. However, optimizing LLMs to hallucinate less on these tasks is challenging, as hallucination is hard to efficiently evaluate at each optimization step. In this work, we show that reducing hallucination on a synthetic task can also reduce hallucination on real-world downstream tasks. Our method, SynTra, first designs a synthetic task where hallucinations are easy to elicit and measure. It next optimizes the LLM's system message via prefix-tuning on the synthetic task, and finally transfers the system message to realistic, hard-to-optimize tasks. Across three realistic abstractive summarization tasks, SynTra reduces hallucination for two 13B-parameter LLMs using only a synthetic retrieval task for supervision. We also find that optimizing the system message rather than the model weights can be critical; fine-tuning the entire model on the synthetic task can counterintuitively increase hallucination. Overall, SynTra demonstrates that the extra flexibility of working with synthetic data can help mitigate undesired behaviors in practice.

Large language models (LLMs) are a new and powerful tool for a wide span of applications involving natural language and demonstrate impressive code generation abilities. The goal of this work is to automatically generate tests and use these tests to validate and verify compiler implementations of a directive-based parallel programming paradigm, OpenACC. To do so, in this paper, we explore the capabilities of state-of-the-art LLMs, including open-source LLMs -- Meta Codellama, Phind fine-tuned version of Codellama, Deepseek Deepseek Coder and closed-source LLMs -- OpenAI GPT-3.5-Turbo and GPT-4-Turbo. We further fine-tuned the open-source LLMs and GPT-3.5-Turbo using our own testsuite dataset along with using the OpenACC specification. We also explored these LLMs using various prompt engineering techniques that include code template, template with retrieval-augmented generation (RAG), one-shot example, one-shot with RAG, expressive prompt with code template and RAG. This paper highlights our findings from over 5000 tests generated via all the above mentioned methods. Our contributions include: (a) exploring the capabilities of the latest and relevant LLMs for code generation, (b) investigating fine-tuning and prompt methods, and (c) analyzing the outcome of LLMs generated tests including manually analysis of representative set of tests. We found the LLM Deepseek-Coder-33b-Instruct produced the most passing tests followed by GPT-4-Turbo.

Machine learning (ML) is increasingly being deployed in critical systems. The data dependence of ML makes securing data used to train and test ML-enabled systems of utmost importance. While the field of cybersecurity has well-established practices for securing information, ML-enabled systems create new attack vectors. Furthermore, data science and cybersecurity domains adhere to their own set of skills and terminologies. This survey aims to present background information for experts in both domains in topics such as cryptography, access control, zero trust architectures, homomorphic encryption, differential privacy for machine learning, and federated learning to establish shared foundations and promote advancements in data security.