This paper addresses the challenging problem of energy-efficient and uncertainty-aware pose estimation in insect-scale drones, which is crucial for tasks such as surveillance in constricted spaces and for enabling non-intrusive spatial intelligence in smart homes. Since tiny drones operate in highly dynamic environments, where factors like lighting and human movement impact their predictive accuracy, it is crucial to deploy uncertainty-aware prediction algorithms that can account for environmental variations and express not only the prediction but also confidence in the prediction. We address both of these challenges with Compute-in-Memory (CIM) which has become a pivotal technology for deep learning acceleration at the edge. While traditional CIM techniques are promising for energy-efficient deep learning, to bring in the robustness of uncertainty-aware predictions at the edge, we introduce a suite of novel techniques: First, we discuss CIM-based acceleration of Bayesian filtering methods uniquely by leveraging the Gaussian-like switching current of CMOS inverters along with co-design of kernel functions to operate with extreme parallelism and with extreme energy efficiency. Secondly, we discuss the CIM-based acceleration of variational inference of deep learning models through probabilistic processing while unfolding iterative computations of the method with a compute reuse strategy to significantly minimize the workload. Overall, our co-design methodologies demonstrate the potential of CIM to improve the processing efficiency of uncertainty-aware algorithms by orders of magnitude, thereby enabling edge robotics to access the robustness of sophisticated prediction frameworks within their extremely stringent area/power resources.

Causal discovery is the challenging task of inferring causal structure from data. Motivated by Pearl's Causal Hierarchy (PCH), which tells us that passive observations alone are not enough to distinguish correlation from causation, there has been a recent push to incorporate interventions into machine learning research. Reinforcement learning provides a convenient framework for such an active approach to learning. This paper presents CORE, a deep reinforcement learning-based approach for causal discovery and intervention planning. CORE learns to sequentially reconstruct causal graphs from data while learning to perform informative interventions. Our results demonstrate that CORE generalizes to unseen graphs and efficiently uncovers causal structures. Furthermore, CORE scales to larger graphs with up to 10 variables and outperforms existing approaches in structure estimation accuracy and sample efficiency. All relevant code and supplementary material can be found at https://github.com/sa-and/CORE

This paper presents a comprehensive study on using deep reinforcement learning (RL) to create dynamic locomotion controllers for bipedal robots. Going beyond focusing on a single locomotion skill, we develop a general control solution that can be used for a range of dynamic bipedal skills, from periodic walking and running to aperiodic jumping and standing. Our RL-based controller incorporates a novel dual-history architecture, utilizing both a long-term and short-term input/output (I/O) history of the robot. This control architecture, when trained through the proposed end-to-end RL approach, consistently outperforms other methods across a diverse range of skills in both simulation and the real world.The study also delves into the adaptivity and robustness introduced by the proposed RL system in developing locomotion controllers. We demonstrate that the proposed architecture can adapt to both time-invariant dynamics shifts and time-variant changes, such as contact events, by effectively using the robot's I/O history. Additionally, we identify task randomization as another key source of robustness, fostering better task generalization and compliance to disturbances. The resulting control policies can be successfully deployed on Cassie, a torque-controlled human-sized bipedal robot. This work pushes the limits of agility for bipedal robots through extensive real-world experiments. We demonstrate a diverse range of locomotion skills, including: robust standing, versatile walking, fast running with a demonstration of a 400-meter dash, and a diverse set of jumping skills, such as standing long jumps and high jumps.

Knowledge Tracing (KT) aims to predict the future performance of students by tracking the development of their knowledge states. Despite all the recent progress made in this field, the application of KT models in education systems is still restricted from the data perspectives: 1) limited access to real life data due to data protection concerns, 2) lack of diversity in public datasets, 3) noises in benchmark datasets such as duplicate records. To resolve these problems, we simulated student data with three statistical strategies based on public datasets and tested their performance on two KT baselines. While we observe only minor performance improvement with additional synthetic data, our work shows that using only synthetic data for training can lead to similar performance as real data.

In this paper, we propose an online algorithm "mspace" for forecasting node features in temporal graphs, which adeptly captures spatial cross-correlation among different nodes as well as the temporal autocorrelation within a node. The algorithm can be used for both probabilistic and deterministic multi-step forecasting, making it applicable for estimation and generation tasks. Comparative evaluations against various baselines, including graph neural network (GNN) based models and classical Kalman filters, demonstrate that mspace performs at par with the state-of-the-art and even surpasses them on some datasets. Importantly, mspace demonstrates consistent robustness across datasets with varying training sizes, a notable advantage over GNN-based methods requiring abundant training samples to learn the spatiotemporal trends in the data effectively. Therefore, employing mspace is advantageous in scenarios where the training sample availability is limited. Additionally, we establish theoretical bounds on multi-step forecasting error of mspace and show that it scales as $O(q)$ for $q$-step forecast.

Training machine learning and statistical models often involves optimizing a data-driven risk criterion. The risk is usually computed with respect to the empirical data distribution, but this may result in poor and unstable out-of-sample performance due to distributional uncertainty. In the spirit of distributionally robust optimization, we propose a novel robust criterion by combining insights from Bayesian nonparametric (i.e., Dirichlet Process) theory and recent decision-theoretic models of smooth ambiguity-averse preferences. First, we highlight novel connections with standard regularized empirical risk minimization techniques, among which Ridge and LASSO regressions. Then, we theoretically demonstrate the existence of favorable finite-sample and asymptotic statistical guarantees on the performance of the robust optimization procedure. For practical implementation, we propose and study tractable approximations of the criterion based on well-known Dirichlet Process representations. We also show that the smoothness of the criterion naturally leads to standard gradient-based numerical optimization. Finally, we provide insights into the workings of our method by applying it to high-dimensional sparse linear regression and robust location parameter estimation tasks.

There are now many explainable AI methods for understanding the decisions of a machine learning model. Among these are those based on counterfactual reasoning, which involve simulating features changes and observing the impact on the prediction. This article proposes to view this simulation process as a source of creating a certain amount of knowledge that can be stored to be used, later, in different ways. This process is illustrated in the additive model and, more specifically, in the case of the naive Bayes classifier, whose interesting properties for this purpose are shown.

Effective communication protocols in multi-agent reinforcement learning (MARL) are critical to fostering cooperation and enhancing team performance. To leverage communication, many previous works have proposed to compress local information into a single message and broadcast it to all reachable agents. This simplistic messaging mechanism, however, may fail to provide adequate, critical, and relevant information to individual agents, especially in severely bandwidth-limited scenarios. This motivates us to develop context-aware communication schemes for MARL, aiming to deliver personalized messages to different agents. Our communication protocol, named CACOM, consists of two stages. In the first stage, agents exchange coarse representations in a broadcast fashion, providing context for the second stage. Following this, agents utilize attention mechanisms in the second stage to selectively generate messages personalized for the receivers. Furthermore, we employ the learned step size quantization (LSQ) technique for message quantization to reduce the communication overhead. To evaluate the effectiveness of CACOM, we integrate it with both actor-critic and value-based MARL algorithms. Empirical results on cooperative benchmark tasks demonstrate that CACOM provides evident performance gains over baselines under communication-constrained scenarios. The code is publicly available at https://github.com/LXXXXR/CACOM.

Active learning (AL) has found wide applications in medical image segmentation, aiming to alleviate the annotation workload and enhance performance. Conventional uncertainty-based AL methods, such as entropy and Bayesian, often rely on an aggregate of all pixel-level metrics. However, in imbalanced settings, these methods tend to neglect the significance of target regions, eg., lesions, and tumors. Moreover, uncertainty-based selection introduces redundancy. These factors lead to unsatisfactory performance, and in many cases, even underperform random sampling. To solve this problem, we introduce a novel approach called the Selective Uncertainty-based AL, avoiding the conventional practice of summing up the metrics of all pixels. Through a filtering process, our strategy prioritizes pixels within target areas and those near decision boundaries. This resolves the aforementioned disregard for target areas and redundancy. Our method showed substantial improvements across five different uncertainty-based methods and two distinct datasets, utilizing fewer labeled data to reach the supervised baseline and consistently achieving the highest overall performance. Our code is available at https://github.com/HelenMa9998/Selective\_Uncertainty\_AL.

The performance of Markov chain Monte Carlo samplers strongly depends on the properties of the target distribution such as its covariance structure, the location of its probability mass and its tail behavior. We explore the use of bijective affine transformations of the sample space to improve the properties of the target distribution and thereby the performance of samplers running in the transformed space. In particular, we propose a flexible and user-friendly scheme for adaptively learning the affine transformation during sampling. Moreover, the combination of our scheme with Gibbsian polar slice sampling is shown to produce samples of high quality at comparatively low computational cost in several settings based on real-world data.