Three major challenges in reinforcement learning are the complex dynamical systems with large state spaces, the costly data acquisition processes, and the deviation of real-world dynamics from the training environment deployment. To overcome these issues, we study distributionally robust Markov decision processes with continuous state spaces under the widely used Kullback-Leibler, chi-square, and total variation uncertainty sets. We propose a model-based approach that utilizes Gaussian Processes and the maximum variance reduction algorithm to efficiently learn multi-output nominal transition dynamics, leveraging access to a generative model (i.e., simulator). We further demonstrate the statistical sample complexity of the proposed method for different uncertainty sets. These complexity bounds are independent of the number of states and extend beyond linear dynamics, ensuring the effectiveness of our approach in identifying near-optimal distributionally-robust policies. The proposed method can be further combined with other model-free distributionally robust reinforcement learning methods to obtain a near-optimal robust policy. Experimental results demonstrate the robustness of our algorithm to distributional shifts and its superior performance in terms of the number of samples needed.

Reinforcement learning policy evaluation problems are often modeled as finite or discounted/averaged infinite-horizon MDPs. In this paper, we study undiscounted off-policy policy evaluation for absorbing MDPs. Given the dataset consisting of the i.i.d episodes with a given truncation level, we propose a so-called MWLA algorithm to directly estimate the expected return via the importance ratio of the state-action occupancy measure. The Mean Square Error (MSE) bound for the MWLA method is investigated and the dependence of statistical errors on the data size and the truncation level are analyzed. With an episodic taxi environment, computational experiments illustrate the performance of the MWLA algorithm.

Modern deep learning heavily relies on large labeled datasets, which often comse with high costs in terms of both manual labeling and computational resources. To mitigate these challenges, researchers have explored the use of informative subset selection techniques, including coreset selection and active learning. Specifically, coreset selection involves sampling data with both input ($\bx$) and output ($\by$), active learning focuses solely on the input data ($\bx$). In this study, we present a theoretically optimal solution for addressing both coreset selection and active learning within the context of linear softmax regression. Our proposed method, COPS (unCertainty based OPtimal Sub-sampling), is designed to minimize the expected loss of a model trained on subsampled data. Unlike existing approaches that rely on explicit calculations of the inverse covariance matrix, which are not easily applicable to deep learning scenarios, COPS leverages the model's logits to estimate the sampling ratio. This sampling ratio is closely associated with model uncertainty and can be effectively applied to deep learning tasks. Furthermore, we address the challenge of model sensitivity to misspecification by incorporating a down-weighting approach for low-density samples, drawing inspiration from previous works. To assess the effectiveness of our proposed method, we conducted extensive empirical experiments using deep neural networks on benchmark datasets. The results consistently showcase the superior performance of COPS compared to baseline methods, reaffirming its efficacy.

Biological sequence analysis relies on the ability to denoise the imprecise output of sequencing platforms. We consider a common setting where a short sequence is read out repeatedly using a high-throughput long-read platform to generate multiple subreads, or noisy observations of the same sequence. Denoising these subreads with alignment-based approaches often fails when too few subreads are available or error rates are too high. In this paper, we propose a novel method for blindly denoising sets of sequences without directly observing clean source sequence labels. Our method, Self-Supervised Set Learning (SSSL), gathers subreads together in an embedding space and estimates a single set embedding as the midpoint of the subreads in both the latent and sequence spaces. This set embedding represents the "average" of the subreads and can be decoded into a prediction of the clean sequence. In experiments on simulated long-read DNA data, SSSL methods denoise small reads of $\leq 6$ subreads with 17% fewer errors and large reads of $>6$ subreads with 8% fewer errors compared to the best baseline. On a real dataset of antibody sequences, SSSL improves over baselines on two self-supervised metrics, with a significant improvement on difficult small reads that comprise over 60% of the test set. By accurately denoising these reads, SSSL promises to better realize the potential of high-throughput DNA sequencing data for downstream scientific applications.

In this paper, we investigate transfer learning in partially observable contextual bandits, where agents have limited knowledge from other agents and partial information about hidden confounders. We first convert the problem to identifying or partially identifying causal effects between actions and rewards through optimization problems. To solve these optimization problems, we discretize the original functional constraints of unknown distributions into linear constraints, and sample compatible causal models via sequentially solving linear programmings to obtain causal bounds with the consideration of estimation error. Our sampling algorithms provide desirable convergence results for suitable sampling distributions. We then show how causal bounds can be applied to improving classical bandit algorithms and affect the regrets with respect to the size of action sets and function spaces. Notably, in the task with function approximation which allows us to handle general context distributions, our method improves the order dependence on function space size compared with previous literatures. We formally prove that our causally enhanced algorithms outperform classical bandit algorithms and achieve orders of magnitude faster convergence rates. Finally, we perform simulations that demonstrate the efficiency of our strategy compared to the current state-of-the-art methods. This research has the potential to enhance the performance of contextual bandit agents in real-world applications where data is scarce and costly to obtain.

Large language models (LLMs) exhibit a wide range of promising capabilities -- from step-by-step planning to commonsense reasoning -- that may provide utility for robots, but remain prone to confidently hallucinated predictions. In this work, we present KnowNo, which is a framework for measuring and aligning the uncertainty of LLM-based planners such that they know when they don't know and ask for help when needed. KnowNo builds on the theory of conformal prediction to provide statistical guarantees on task completion while minimizing human help in complex multi-step planning settings. Experiments across a variety of simulated and real robot setups that involve tasks with different modes of ambiguity (e.g., from spatial to numeric uncertainties, from human preferences to Winograd schemas) show that KnowNo performs favorably over modern baselines (which may involve ensembles or extensive prompt tuning) in terms of improving efficiency and autonomy, while providing formal assurances. KnowNo can be used with LLMs out of the box without model-finetuning, and suggests a promising lightweight approach to modeling uncertainty that can complement and scale with the growing capabilities of foundation models. Website: https://robot-help.github.io

Following four successful years in the SAE AutoDrive Challenge Series I, the University of Toronto is participating in the Series II competition to develop a Level 4 autonomous passenger vehicle capable of handling various urban driving scenarios by 2025. Accurate detection of traffic lights and correct identification of their states is essential for safe autonomous operation in cities. Herein, we describe our recently-redesigned traffic light perception system for autonomous vehicles like the University of Toronto's self-driving car, Artemis. Similar to most traffic light perception systems, we rely primarily on camera-based object detectors. We deploy the YOLOv5 detector for bounding box regression and traffic light classification across multiple cameras and fuse the observations. To improve robustness, we incorporate priors from high-definition semantic maps and perform state filtering using hidden Markov models. We demonstrate a multi-camera, real time-capable traffic light perception pipeline that handles complex situations including multiple visible intersections, traffic light variations, temporary occlusion, and flashing light states. To validate our system, we collected and annotated a varied dataset incorporating flashing states and a range of occlusion types. Our results show superior performance in challenging real-world scenarios compared to single-frame, single-camera object detection.

Global fits of physics models require efficient methods for exploring high-dimensional and/or multimodal posterior functions. We introduce a novel method for accelerating Markov Chain Monte Carlo (MCMC) sampling by pairing a Metropolis-Hastings algorithm with a diffusion model that can draw global samples with the aim of approximating the posterior. We briefly review diffusion models in the context of image synthesis before providing a streamlined diffusion model tailored towards low-dimensional data arrays. We then present our adapted Metropolis-Hastings algorithm which combines local proposals with global proposals taken from a diffusion model that is regularly trained on the samples produced during the MCMC run. Our approach leads to a significant reduction in the number of likelihood evaluations required to obtain an accurate representation of the Bayesian posterior across several analytic functions, as well as for a physical example based on a global analysis of parton distribution functions. Our method is extensible to other MCMC techniques, and we briefly compare our method to similar approaches based on normalizing flows. A code implementation can be found at https://github.com/NickHunt-Smith/MCMC-diffusion.

Categorical sequence clustering plays a crucial role in various fields, but the lack of interpretability in cluster assignments poses significant challenges. Sequences inherently lack explicit features, and existing sequence clustering algorithms heavily rely on complex representations, making it difficult to explain their results. To address this issue, we propose a method called Interpretable Sequence Clustering Tree (ISCT), which combines sequential patterns with a concise and interpretable tree structure. ISCT leverages k-1 patterns to generate k leaf nodes, corresponding to k clusters, which provides an intuitive explanation on how each cluster is formed. More precisely, ISCT first projects sequences into random subspaces and then utilizes the k-means algorithm to obtain high-quality initial cluster assignments. Subsequently, it constructs a pattern-based decision tree using a boosting-based construction strategy in which sequences are re-projected and re-clustered at each node before mining the top-1 discriminative splitting pattern. Experimental results on 14 real-world data sets demonstrate that our proposed method provides an interpretable tree structure while delivering fast and accurate cluster assignments.

Bruna Moreira da Silva is a PhD student at The University of Melbourne. Her research interests are in bioinformatics, immunoinformatics and machine learning to advance Global Health. David B. Ascher is the Director of Biotechnology at the University of Queensland and head of Computational Biology and Clinical Informatics at the Baker Institute and Systems and Computational Biology at Bio21 Institute. He is interested in developing and applying computational tools to assist leveraging clinical and omics data for drug discovery and personalised medicine. Nicholas Geard is an Associate Professor at the School of Computing and Information Systems at the University of Melbourne and Director of the Melbourne Data Analytics Platform. He is a computer scientist specialising in computational simulation applied to a range of problems in health and epidemiology. Douglas E. V. Pires is an Associate Professor in Digital Health at the School of Computing and Information Systems at the University of Melbourne and group leader at the Bio21 Institute. He is a computer scientist and bioinformatician specialising in machine learning and AI and the development of the next generation of tools to analyse omics data, and guide drug discovery and personalised medicine. ABSTRACT Antibodies, a prominent class of approved biologics, play a crucial role in detecting foreign antigens. The effectiveness of antigen neutralisation and elimination hinges upon the strength, sensitivity, and specificity of the paratope-epitope interaction, which demands resource-intensive experimental techniques for characterisation. In recent years, artificial intelligence and machine learning methods have made significant strides, revolutionising the prediction of protein structures and their complexes. The past decade has also witnessed the evolution of computational approaches aiming to support immunotherapy design. This review focuses on the progress of machine learning-based tools and their frameworks in the domain of B-cell immunotherapy design, encompassing linear and conformational epitope prediction, paratope prediction, and antibody design. We mapped the most commonly used data sources, evaluation metrics, and method availability and thoroughly assessed their significance and limitations, discussing the main challenges ahead. INTRODUCTION Therapeutic antibodies are a rapidly growing class of biopharmaceuticals with potentially exceptional antigen specificity and affinity. Their ability to detect and eliminate a wide array of foreign threats makes them suitable for a range of potential therapeutic and diagnostic applications. Antibody and antigen engineering have been greatly benefited by the evolution of research in computational biology, leading to innovative approaches in screening antibody targets, optimising their biochemical and physical properties, predicting and optimising binding affinity and understanding escape mutations [1].