We present a new computing model for intrinsic rewards in reinforcement learning that addresses the limitations of existing surprise-driven explorations. The reward is the novelty of the surprise rather than the surprise norm. We estimate the surprise novelty as retrieval errors of a memory network wherein the memory stores and reconstructs surprises. Our surprise memory (SM) augments the capability of surprise-based intrinsic motivators, maintaining the agent's interest in exciting exploration while reducing unwanted attraction to unpredictable or noisy observations. Our experiments demonstrate that the SM combined with various surprise predictors exhibits efficient exploring behaviors and significantly boosts the final performance in sparse reward environments, including Noisy-TV, navigation and challenging Atari games.

We address the task of probabilistic anomaly attribution in the black-box regression setting, where the goal is to compute the probability distribution of the attribution score of each input variable, given an observed anomaly. The training dataset is assumed to be unavailable. This task differs from the standard XAI (explainable AI) scenario, since we wish to explain the anomalous deviation from a black-box prediction rather than the black-box model itself. We begin by showing that mainstream model-agnostic explanation methods, such as the Shapley values, are not suitable for this task because of their ``deviation-agnostic property.'' We then propose a novel framework for probabilistic anomaly attribution that allows us to not only compute attribution scores as the predictive mean but also quantify the uncertainty of those scores. This is done by considering a generative process for perturbations that counter-factually bring the observed anomalous observation back to normalcy. We introduce a variational Bayes algorithm for deriving the distributions of per variable attribution scores. To the best of our knowledge, this is the first probabilistic anomaly attribution framework that is free from being deviation-agnostic.

This comprehensive review systematically evaluates Machine Learning (ML) methodologies employed in the detection, prediction, and analysis of mental stress and its consequent mental disorders (MDs). Utilizing a rigorous scoping review process, the investigation delves into the latest ML algorithms, preprocessing techniques, and data types employed in the context of stress and stress-related MDs. The findings highlight that Support Vector Machine (SVM), Neural Network (NN), and Random Forest (RF) models consistently exhibit superior accuracy and robustness among all machine learning algorithms examined. Furthermore, the review underscores that physiological parameters, such as heart rate measurements and skin response, are prevalently used as stress predictors in ML algorithms. This is attributed to their rich explanatory information concerning stress and stress-related MDs, as well as the relative ease of data acquisition. Additionally, the application of dimensionality reduction techniques, including mappings, feature selection, filtering, and noise reduction, is frequently observed as a crucial step preceding the training of ML algorithms. The synthesis of this review identifies significant research gaps and outlines future directions for the field. These encompass areas such as model interpretability, model personalization, the incorporation of naturalistic settings, and real-time processing capabilities for detection and prediction of stress and stress-related MDs.

This paper analyzes finite state Markov Decision Processes (MDPs) with uncertain parameters in compact sets and re-examines results from robust MDP via set-based fixed point theory. To this end, we generalize the Bellman and policy evaluation operators to contracting operators on the value function space and denote them as \emph{value operators}. We lift these value operators to act on \emph{sets} of value functions and denote them as \emph{set-based value operators}. We prove that the set-based value operators are \emph{contractions} in the space of compact value function sets. Leveraging insights from set theory, we generalize the rectangularity condition in classic robust MDP literature to a containment condition for all value operators, which is weaker and can be applied to a larger set of parameter-uncertain MDPs and contracting operators in dynamic programming. We prove that both the rectangularity condition and the containment condition sufficiently ensure that the set-based value operator's fixed point set contains its own extrema elements. For convex and compact sets of uncertain MDP parameters, we show equivalence between the classic robust value function and the supremum of the fixed point set of the set-based Bellman operator. Under dynamically changing MDP parameters in compact sets, we prove a set convergence result for value iteration, which otherwise may not converge to a single value function. Finally, we derive novel guarantees for probabilistic path-planning problems in planet exploration and stratospheric station-keeping.

Machine learning has achieved remarkable success over the past couple of decades, often attributed to a combination of algorithmic innovations and the availability of high-quality data available at scale. However, a third critical component is the fine-tuning of hyperparameters, which plays a pivotal role in achieving optimal model performance. Despite its significance, hyperparameter optimization (HPO) remains a challenging task for several reasons. Many HPO techniques rely on naive search methods or assume that the loss function is smooth and continuous, which may not always be the case. Traditional methods, like grid search and Bayesian optimization, often struggle to quickly adapt and efficiently search the loss landscape. Grid search is computationally expensive, while Bayesian optimization can be slow to prime. Since the search space for HPO is frequently high-dimensional and non-convex, it is often challenging to efficiently find a global minimum. Moreover, optimal hyperparameters can be sensitive to the specific dataset or task, further complicating the search process. To address these issues, we propose a new hyperparameter optimization method, HomOpt, using a data-driven approach based on a generalized additive model (GAM) surrogate combined with homotopy optimization. This strategy augments established optimization methodologies to boost the performance and effectiveness of any given method with faster convergence to the optimum on continuous, discrete, and categorical domain spaces. We compare the effectiveness of HomOpt applied to multiple optimization techniques (e.g., Random Search, TPE, Bayes, and SMAC) showing improved objective performance on many standardized machine learning benchmarks and challenging open-set recognition tasks.

In real-world scenarios, human dialogues are multi-round and diverse. Furthermore, human instructions can be unclear and human responses are unrestricted. Interactive robots face difficulties in understanding human intents and generating suitable strategies for assisting individuals through manipulation. In this article, we propose Mani-GPT, a Generative Pre-trained Transformer (GPT) for interactive robotic manipulation. The proposed model has the ability to understand the environment through object information, understand human intent through dialogues, generate natural language responses to human input, and generate appropriate manipulation plans to assist the human. This makes the human-robot interaction more natural and humanized. In our experiment, Mani-GPT outperforms existing algorithms with an accuracy of 84.6% in intent recognition and decision-making for actions. Furthermore, it demonstrates satisfying performance in real-world dialogue tests with users, achieving an average response accuracy of 70%.

There is a growing interest in using reinforcement learning (RL) to personalize sequences of treatments in digital health to support users in adopting healthier behaviors. Such sequential decision-making problems involve decisions about when to treat and how to treat based on the user's context (e.g., prior activity level, location, etc.). Online RL is a promising data-driven approach for this problem as it learns based on each user's historical responses and uses that knowledge to personalize these decisions. However, to decide whether the RL algorithm should be included in an ``optimized'' intervention for real-world deployment, we must assess the data evidence indicating that the RL algorithm is actually personalizing the treatments to its users. Due to the stochasticity in the RL algorithm, one may get a false impression that it is learning in certain states and using this learning to provide specific treatments. We use a working definition of personalization and introduce a resampling-based methodology for investigating whether the personalization exhibited by the RL algorithm is an artifact of the RL algorithm stochasticity. We illustrate our methodology with a case study by analyzing the data from a physical activity clinical trial called HeartSteps, which included the use of an online RL algorithm. We demonstrate how our approach enhances data-driven truth-in-advertising of algorithm personalization both across all users as well as within specific users in the study.

When performing causal discovery, assumptions have to be made on how the true causal mechanism corresponds to the underlying joint probability distribution. These assumptions are labeled as causal razors in this work. We review numerous causal razors that appeared in the literature, and offer a comprehensive logical comparison of them. In particular, we scrutinize an unpopular causal razor, namely parameter minimality, in multinomial causal models and its logical relations with other well-studied causal razors. Our logical result poses a dilemma in selecting a reasonable scoring criterion for score-based casual search algorithms.

Infectious diseases, either emerging or long-lasting, place numerous people at risk and bring heavy public health burdens worldwide. In the process against infectious diseases, predicting the epidemic risk by modeling the disease transmission plays an essential role in assisting with preventing and controlling disease transmission in a more effective way. In this paper, we systematically describe how machine learning can play an essential role in quantitatively characterizing disease transmission patterns and accurately predicting infectious disease risks. First, we introduce the background and motivation of using machine learning for infectious disease risk prediction. Next, we describe the development and components of various machine learning models for infectious disease risk prediction. Specifically, existing models fall into three categories: Statistical prediction, data-driven machine learning, and epidemiology-inspired machine learning. Subsequently, we discuss challenges encountered when dealing with model inputs, designing task-oriented objectives, and conducting performance evaluation. Finally, we conclude with a discussion of open questions and future directions.

Non-stationarity is a fundamental challenge in multi-agent reinforcement learning (MARL), where agents update their behaviour as they learn. Many theoretical advances in MARL avoid the challenge of non-stationarity by coordinating the policy updates of agents in various ways, including synchronizing times at which agents are allowed to revise their policies. Synchronization enables analysis of many MARL algorithms via multi-timescale methods, but such synchrony is infeasible in many decentralized applications. In this paper, we study an asynchronous variant of the decentralized Q-learning algorithm, a recent MARL algorithm for stochastic games. We provide sufficient conditions under which the asynchronous algorithm drives play to equilibrium with high probability. Our solution utilizes constant learning rates in the Q-factor update, which we show to be critical for relaxing the synchrony assumptions of earlier work. Our analysis also applies to asynchronous generalizations of a number of other algorithms from the regret testing tradition, whose performance is analyzed by multi-timescale methods that study Markov chains obtained via policy update dynamics. This work extends the applicability of the decentralized Q-learning algorithm and its relatives to settings in which parameters are selected in an independent manner, and tames non-stationarity without imposing the coordination assumptions of prior work.