Communication enables agents to cooperate to achieve their goals. Learning when to communicate, i.e. sparse communication, is particularly important where bandwidth is limited, in situations where agents interact with humans, in partially observable scenarios where agents must convey information unavailable to others, and in non-cooperative scenarios where agents may hide information to gain a competitive advantage. Recent work in learning sparse communication, however, suffers from high variance training where, the price of decreasing communication is a decrease in reward, particularly in cooperative tasks. Sparse communications are necessary to match agent communication to limited human bandwidth. Humans additionally communicate via discrete linguistic tokens, previously shown to decrease task performance when compared to continuous communication vectors. This research addresses the above issues by limiting the loss in reward of decreasing communication and eliminating the penalty for discretization. In this work, we successfully constrain training using a learned gate to regulate when to communicate while using discrete prototypes that reflect what to communicate for cooperative tasks with partial observability. We provide two types of "Enforcers" for hard and soft budget constraints and present results of communication under different budgets. We show that our method satisfies constraints while yielding the same performance as comparable, unconstrained methods.
相關內容
The reinforcement learning (RL) problem is rife with sources of non-stationarity, making it a notoriously difficult problem domain for the application of neural networks. We identify a mechanism by which non-stationary prediction targets can prevent learning progress in deep RL agents: \textit{capacity loss}, whereby networks trained on a sequence of target values lose their ability to quickly update their predictions over time. We demonstrate that capacity loss occurs in a range of RL agents and environments, and is particularly damaging to performance in sparse-reward tasks. We then present a simple regularizer, Initial Feature Regularization (InFeR), that mitigates this phenomenon by regressing a subspace of features towards its value at initialization, leading to significant performance improvements in sparse-reward environments such as Montezuma's Revenge. We conclude that preventing capacity loss is crucial to enable agents to maximally benefit from the learning signals they obtain throughout the entire training trajectory.
We study a new two-time-scale stochastic gradient method for solving optimization problems, where the gradients are computed with the aid of an auxiliary variable under samples generated by time-varying Markov random processes parameterized by the underlying optimization variable. These time-varying samples make gradient directions in our update biased and dependent, which can potentially lead to the divergence of the iterates. In our two-time-scale approach, one scale is to estimate the true gradient from these samples, which is then used to update the estimate of the optimal solution. While these two iterates are implemented simultaneously, the former is updated "faster" (using bigger step sizes) than the latter (using smaller step sizes). Our first contribution is to characterize the finite-time complexity of the proposed two-time-scale stochastic gradient method. In particular, we provide explicit formulas for the convergence rates of this method under different structural assumptions, namely, strong convexity, convexity, the Polyak-Lojasiewicz condition, and general non-convexity. We apply our framework to two problems in control and reinforcement learning. First, we look at the standard online actor-critic algorithm over finite state and action spaces and derive a convergence rate of O(k^(-2/5)), which recovers the best known rate derived specifically for this problem. Second, we study an online actor-critic algorithm for the linear-quadratic regulator and show that a convergence rate of O(k^(-2/3)) is achieved. This is the first time such a result is known in the literature. Finally, we support our theoretical analysis with numerical simulations where the convergence rates are visualized.
Since 2010, the output of a risk assessment tool that predicts how likely an individual is to commit severe violence against their partner has been integrated within the Basque country courtrooms. The EPV-R, the tool developed to assist police officers during the assessment of gender-based violence cases, was also incorporated to assist the decision-making of judges. With insufficient training, judges are exposed to an algorithmic output that influences the human decision of adopting measures in cases of gender-based violence. In this paper, we examine the risks, harms and limits of algorithmic governance within the context of gender-based violence. Through the lens of an Spanish judge exposed to this tool, we analyse how the EPV-R is impacting on the justice system. Moving beyond the risks of unfair and biased algorithmic outputs, we examine legal, social and technical pitfalls such as opaque implementation, efficiency's paradox and feedback loop, that could led to unintended consequences on women who suffer gender-based violence. Our interdisciplinary framework highlights the importance of understanding the impact and influence of risk assessment tools within judicial decision-making and increase awareness about its implementation in this context.
Women are influential online, especially in image-based social media such as Twitter and Instagram. However, many in the network environment contain gender discrimination and aggressive information, which magnify gender stereotypes and gender inequality. Therefore, the filtering of illegal content such as gender discrimination is essential to maintain a healthy social network environment. In this paper, we describe the system developed by our team for SemEval-2022 Task 5: Multimedia Automatic Misogyny Identification. More specifically, we introduce two novel system to analyze these posts: a multimodal multi-task learning architecture that combines Bertweet for text encoding with ResNet-18 for image representation, and a single-flow transformer structure which combines text embeddings from BERT-Embeddings and image embeddings from several different modules such as EfficientNet and ResNet. In this manner, we show that the information behind them can be properly revealed. Our approach achieves good performance on each of the two subtasks of the current competition, ranking 15th for Subtask A (0.746 macro F1-score), 11th for Subtask B (0.706 macro F1-score) while exceeding the official baseline results by high margins.
Integrated sensing and communication (ISAC) creates a platform to exploit the synergy between two powerful functionalities that have been developing separately. However, the interference management and resource allocation between sensing and communication have not been fully studied. In this paper, we consider the design of perceptive mobile networks (PMNs) by adding sensing capability to current cellular networks. To avoid the full-duplex operation, we propose the PMN with distributed target monitoring terminals (TMTs) where passive TMTs are deployed over wireless networks to locate the sensing target (ST). We jointly optimize the transmit and receive beamformers towards the communication user terminals (UEs) and the ST by alternating-optimization (AO) and prove its convergence. To reduce computation complexity and obtain physical insights, we further investigate the use of linear transceivers, including zero forcing and beam synthesis (B-syn). Our analysis revealed interesting physical insights regarding interference management and resource allocation between sensing and communication: 1) instead of forming dedicated sensing signals, it is more efficient to redesign the communication signals for both communication and sensing purposes and "leak" communication energy for sensing; 2) the amount of energy leakage from one UE to the ST depends on their relative locations.
Spectral efficiency improvement is a key focus in most wireless communication systems and achieved by various means such as using large antenna arrays and/or advanced modulation schemes and signal formats. This work proposes to further improve spectral efficiency through combining non-orthogonal spectrally efficient frequency division multiplexing (SEFDM) systems with index modulation (IM), which can efficiently make use of the indices of activated subcarriers as communication information. Recent research has verified that IM may be used with SEFDM to alleviate inter-carrier interference (ICI) and improve error performance. This work proposes new SEFDM signal formats based on novel activation pattern designs, which limit the locations of activated subcarriers and enable a variable number of activated subcarriers in each SEFDM subblock. SEFDM-IM system designs are developed by jointly considering activation patterns, modulation schemes and signal waveform formats, with a set of solutions evaluated under different spectral efficiency scenarios. Detailed modelling of coded systems and simulation studies reveal that the proposed designs not only lead to better bit error rate (BER) but also lower peak-to-average power ratio (PAPR) and reduced computational complexity relative to other reported index-modulated systems.
We study the performance of a phase-noise impaired double reconfigurable intelligent surface (RIS)-aided multiuser (MU) multiple-input single-output (MISO) system under spatial correlation at both RISs and base-station (BS). The downlink achievable rate is derived in closed-form under maximum ratio transmission (MRT) precoding. In addition, we obtain the optimal phase-shift design at both RISs in closed-form for the considered channel and phase-noise models. Numerical results validate the analytical expressions, and highlight the effects of different system parameters on the achievable rate. Our analysis shows that phase-noise can severely degrade the performance when users do not have direct links to both RISs, and can only be served via the double-reflection link. Also, we show that high spatial correlation at RISs is essential for high achievable rates.
Federated learning (FL) has been recognized as a viable distributed learning paradigm which trains a machine learning model collaboratively with massive mobile devices in the wireless edge while protecting user privacy. Although various communication schemes have been proposed to expedite the FL process, most of them have assumed ideal wireless channels which provide reliable and lossless communication links between the server and mobile clients. Unfortunately, in practical systems with limited radio resources such as constraint on the training latency and constraints on the transmission power and bandwidth, transmission of a large number of model parameters inevitably suffers from quantization errors (QE) and transmission outage (TO). In this paper, we consider such non-ideal wireless channels, and carry out the first analysis showing that the FL convergence can be severely jeopardized by TO and QE, but intriguingly can be alleviated if the clients have uniform outage probabilities. These insightful results motivate us to propose a robust FL scheme, named FedTOE, which performs joint allocation of wireless resources and quantization bits across the clients to minimize the QE while making the clients have the same TO probability. Extensive experimental results are presented to show the superior performance of FedTOE for deep learning-based classification tasks with transmission latency constraints.
Multi-camera vehicle tracking is one of the most complicated tasks in Computer Vision as it involves distinct tasks including Vehicle Detection, Tracking, and Re-identification. Despite the challenges, multi-camera vehicle tracking has immense potential in transportation applications including speed, volume, origin-destination (O-D), and routing data generation. Several recent works have addressed the multi-camera tracking problem. However, most of the effort has gone towards improving accuracy on high-quality benchmark datasets while disregarding lower camera resolutions, compression artifacts and the overwhelming amount of computational power and time needed to carry out this task on its edge and thus making it prohibitive for large-scale and real-time deployment. Therefore, in this work we shed light on practical issues that should be addressed for the design of a multi-camera tracking system to provide actionable and timely insights. Moreover, we propose a real-time city-scale multi-camera vehicle tracking system that compares favorably to computationally intensive alternatives and handles real-world, low-resolution CCTV instead of idealized and curated video streams. To show its effectiveness, in addition to integration into the Regional Integrated Transportation Information System (RITIS), we participated in the 2021 NVIDIA AI City multi-camera tracking challenge and our method is ranked among the top five performers on the public leaderboard.
Effective multi-robot teams require the ability to move to goals in complex environments in order to address real-world applications such as search and rescue. Multi-robot teams should be able to operate in a completely decentralized manner, with individual robot team members being capable of acting without explicit communication between neighbors. In this paper, we propose a novel game theoretic model that enables decentralized and communication-free navigation to a goal position. Robots each play their own distributed game by estimating the behavior of their local teammates in order to identify behaviors that move them in the direction of the goal, while also avoiding obstacles and maintaining team cohesion without collisions. We prove theoretically that generated actions approach a Nash equilibrium, which also corresponds to an optimal strategy identified for each robot. We show through extensive simulations that our approach enables decentralized and communication-free navigation by a multi-robot system to a goal position, and is able to avoid obstacles and collisions, maintain connectivity, and respond robustly to sensor noise.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191