Thanks to the explosive developments of data-driven learning methodologies recently, reinforcement learning (RL) emerges as a promising solution to address the legged locomotion problem in robotics. In this manuscript, we propose a novel concurrent teacher-student reinforcement learning architecture for legged locomotion over challenging terrains, based only on proprioceptive measurements in real-world deployment. Different from convectional teacher-student architecture that trains the teacher policy via RL and transfers the knowledge to the student policy through supervised learning, our proposed architecture trains teacher and student policy networks concurrently under the reinforcement learning paradigm. To achieve this, we develop a new training scheme based on conventional proximal policy gradient (PPO) method to accommodate the interaction between teacher policy network and student policy network. The effectiveness of the proposed architecture as well as the new training scheme is demonstrated through extensive indoor and outdoor experiments on quadrupedal robots and point-foot bipedal robot, showcasing robust locomotion over challenging terrains and improved performance compared to two-stage training methods.

Sleep is vital for people's physical and mental health, and sound sleep can help them focus on daily activities. Therefore, a sleep study that includes sleep patterns and disorders is crucial to enhancing our knowledge about individuals' health status. The findings on sleep stages and sleep disorders relied on polysomnography and self-report measures, and then the study went through clinical assessments by expert physicians. However, the evaluation process of sleep stage classification and sleep disorder has become more convenient with artificial intelligence applications and numerous investigations focusing on various datasets with advanced algorithms and techniques that offer improved computational ease and accuracy. This study aims to provide a comprehensive, systematic review and meta-analysis of the recent literature to analyze the different approaches and their outcomes in sleep studies, which includes works on sleep stages classification and sleep disorder detection using AI. In this review, 183 articles were initially selected from different journals, among which 80 records were enlisted for explicit review, ranging from 2016 to 2023. Brain waves were the most commonly employed body parameters for sleep staging and disorder studies. The convolutional neural network, the most widely used of the 34 distinct artificial intelligence models, comprised 27%. The other models included the long short-term memory, support vector machine, random forest, and recurrent neural network, which consisted of 11%, 6%, 6%, and 5% sequentially. For performance metrics, accuracy was widely used for a maximum of 83.75% of the cases, the F1 score of 45%, Kappa of 36.25%, Sensitivity of 31.25%, and Specificity of 30% of cases, along with the other metrics. This article would help physicians and researchers get the gist of AI's contribution to sleep studies and the feasibility of their intended work.

Unmanned Aerial Vehicles (UAVs) have gained popularity in data harvesting (DH) and coverage path planning (CPP) to survey a given area efficiently and collect data from aerial perspectives, while data harvesting aims to gather information from various Internet of Things (IoT) sensor devices, coverage path planning guarantees that every location within the designated area is visited with minimal redundancy and maximum efficiency. We propose the ARDDQN (Attention-based Recurrent Double Deep Q Network), which integrates double deep Q-networks (DDQN) with recurrent neural networks (RNNs) and an attention mechanism to generate path coverage choices that maximize data collection from IoT devices and to learn a control scheme for the UAV that generalizes energy restrictions. We employ a structured environment map comprising a compressed global environment map and a local map showing the UAV agent's locate efficiently scaling to large environments. We have compared Long short-term memory (LSTM), Bi-directional long short-term memory (Bi-LSTM), Gated recurrent unit (GRU) and Bidirectional gated recurrent unit (Bi-GRU) as recurrent neural networks (RNN) to the result without RNN We propose integrating the LSTM with the Attention mechanism to the existing DDQN model, which works best on evolution parameters, i.e., data collection, landing, and coverage ratios for the CPP and data harvesting scenarios.

Over the last few years, 360{\deg} video traffic on the network has grown significantly. A key challenge of 360{\deg} video playback is ensuring a high quality of experience (QoE) with limited network bandwidth. Currently, most studies focus on tile-based adaptive bitrate (ABR) streaming based on single viewport prediction to reduce bandwidth consumption. However, the performance of models for single-viewpoint prediction is severely limited by the inherent uncertainty in head movement, which can not cope with the sudden movement of users very well. This paper first presents a multimodal spatial-temporal attention transformer to generate multiple viewpoint trajectories with their probabilities given a historical trajectory. The proposed method models viewpoint prediction as a classification problem and uses attention mechanisms to capture the spatial and temporal characteristics of input video frames and viewpoint trajectories for multi-viewpoint prediction. After that, a multi-agent deep reinforcement learning (MADRL)-based ABR algorithm utilizing multi-viewpoint prediction for 360{\deg} video streaming is proposed for maximizing different QoE objectives under various network conditions. We formulate the ABR problem as a decentralized partially observable Markov decision process (Dec-POMDP) problem and present a MAPPO algorithm based on centralized training and decentralized execution (CTDE) framework to solve the problem. The experimental results show that our proposed method improves the defined QoE metric by up to 85.5% compared to existing ABR methods.

Generative models aim to create samples statistically similar to those belonging to a training dataset: their goal is to fit the probability distribution from which the datapoints supposedly come from. In a generic setting, this probability distribution takes the form of a Boltzmann factor. The corresponding Energy Based Models (EBMs) fit the parameters of the Hamiltonian of the Boltzmann distribution and can be viewed as maximum entropy models, where the statistical properties of the dataset are imposed as constraints to low degree correlation functions [1, 2], (see [3, 4] for recent reviews). The resulting learning rule can be viewed as a gradient ascent on the Log-Likelihood (LL). However, running the training dynamics is a notoriously challenging task: at each epoch, the evaluation of the gradient of the LL requires the computation of the correlation functions of the degrees of freedom as predicted from the current estimation of the model's probability distribution. This is typically an intractable problem from an analytical point of view and is generally tackled numerically through parallel Monte Carlo Markov Chain (MCMC) simulations.

Pragmatic or goal-oriented communication can optimize communication decisions beyond the reliable transmission of data, instead aiming at directly affecting application performance with the minimum channel utilization. In this paper, we develop a general theoretical framework for the remote control of finite-state Markov processes, using pragmatic communication over a costly zero-delay communication channel. To that end, we model a cyber-physical system composed of an encoder, which observes and transmits the states of a process in real-time, and a decoder, which receives that information and controls the behavior of the process. The encoder and the decoder should cooperatively optimize the trade-off between the control performance (i.e., reward) and the communication cost (i.e., channel use). This scenario underscores a pragmatic (i.e., goal-oriented) communication problem, where the purpose is to convey only the data that is most valuable for the underlying task, taking into account the state of the decoder (hence, the pragmatic aspect). We investigate two different decision-making architectures: in pull-based remote control, the decoder is the only decision-maker, while in push-based remote control, the encoder and the decoder constitute two independent decision-makers, leading to a multi-agent scenario. We propose three algorithms to optimize our system (i.e., design the encoder and the decoder policies), discuss the optimality guarantees ofs the algorithms, and shed light on their computational complexity and fundamental limits.

In recent years, Large Language Models (LLMs) have shown great abilities in various tasks, including question answering, arithmetic problem solving, and poem writing, among others. Although research on LLM-as-an-agent has shown that LLM can be applied to Reinforcement Learning (RL) and achieve decent results, the extension of LLM-based RL to Multi-Agent System (MAS) is not trivial, as many aspects, such as coordination and communication between agents, are not considered in the RL frameworks of a single agent. To inspire more research on LLM-based MARL, in this letter, we survey the existing LLM-based single-agent and multi-agent RL frameworks and provide potential research directions for future research. In particular, we focus on the cooperative tasks of multiple agents with a common goal and communication among them. We also consider human-in/on-the-loop scenarios enabled by the language component in the framework.

In this paper, we introduce the first comprehensive multilingual sign language dataset named Prompt2Sign, which builds from public data including American Sign Language (ASL) and seven others. Our dataset transforms a vast array of videos into a streamlined, model-friendly format, optimized for training with translation models like seq2seq and text2text. Building on this new dataset, we propose SignLLM, the first multilingual Sign Language Production (SLP) model, which includes two novel multilingual SLP modes that allow for the generation of sign language gestures from input text or prompt. Both of the modes can use a new loss and a module based on reinforcement learning, which accelerates the training by enhancing the model's capability to autonomously sample high-quality data. We present benchmark results of SignLLM, which demonstrate that our model achieves state-of-the-art performance on SLP tasks across eight sign languages. More code and materials are available at https://signllm.github.io/.

A common problem for agents operating in real-world environments is that the response of an environment to their actions may be non-deterministic and observed through noise. This renders environmental state and progress towards completing a task latent. Despite recent impressive demonstrations of LLM's reasoning abilities on various benchmarks, whether LLMs can build estimates of latent state and leverage them for reasoning has not been explicitly studied. We investigate this problem in the real-world domain of autonomous UI agents. We establish that appropriately prompting LLMs in a zero-shot manner can be formally understood as forming point estimates of latent state in a textual space. In the context of autonomous UI agents we then show that LLMs used in this manner are more than $76\%$ accurate at inferring various aspects of latent state, such as performed (vs. commanded) actions and task progression. Using both public and internal benchmarks and three reasoning methods (zero-shot, CoT-SC & ReAct), we show that LLM-powered agents that explicitly estimate and reason about latent state are able to successfully complete up to 1.6x more tasks than those that do not.

We model a Markov decision process, parametrized by an unknown parameter, and study the asymptotic behavior of a sampling-based algorithm, called Thompson sampling. The standard definition of regret is not always suitable to evaluate a policy, especially when the underlying chain structure is general. We show that the standard (expected) regret can grow (super-)linearly and fails to capture the notion of learning in realistic settings with non-trivial state evolution. By decomposing the standard (expected) regret, we develop a new metric, called the expected residual regret, which forgets the immutable consequences of past actions. Instead, it measures regret against the optimal reward moving forward from the current period. We show that the expected residual regret of the Thompson sampling algorithm is upper bounded by a term which converges exponentially fast to 0. We present conditions under which the posterior sampling error of Thompson sampling converges to 0 almost surely. We then introduce the probabilistic version of the expected residual regret and present conditions under which it converges to 0 almost surely. Thus, we provide a viable concept of learning for sampling algorithms which will serve useful in broader settings than had been considered previously.