Although perception is an increasingly dominant portion of the overall computational cost for autonomous systems, only a fraction of the information perceived is likely to be relevant to the current task. To alleviate these perception costs, we develop a novel simultaneous perception-action design framework wherein an agent senses only the task-relevant information. This formulation differs from that of a partially observable Markov decision process, since the agent is free to synthesize not only its policy for action selection but also its belief-dependent observation function. The method enables the agent to balance its perception costs with those incurred by operating in its environment. To obtain a computationally tractable solution, we approximate the value function using a novel method of invariant finite belief sets, wherein the agent acts exclusively on a finite subset of the continuous belief space. We solve the approximate problem through value iteration in which a linear program is solved individually for each belief state in the set, in each iteration. Finally, we prove that the value functions, under an assumption on their structure, converge to their continuous state-space values as the sample density increases.
Gap the (Theory of) Mind: Sharing Beliefs About Teammates' Goals Boosts Collaboration Perception, Not Performance
Faculty of Data & Decisions Science Technion - Israel Institute of Technology（数据与决策科学学院，技术学院-以色列理工学院）；Department of Computer Science Tufts University（计算机科学系，塔夫茨大学）
Massachusetts Institute of Technology (MIT)（麻省理工学院）
We pose an active perception problem where an autonomous agent actively interacts with a second agent with potentially adversarial behaviors. Given the uncertainty in the intent of the other agent, the objective is to collect further evidence to help discriminate potential threats. The main technical challenges are the partial observability of the agent intent, the adversary modeling, and the corresponding uncertainty modeling. Note that an adversary agent may act to mislead the autonomous agent by using a deceptive strategy that is learned from past experiences. We propose an approach that co…
University of California at Berkeley（加州大学伯克利分校）；California PATH, University of California at Berkeley（加州大学伯克利分校加州PATH）
Autonomous cars have to navigate in dynamic environment which can be full of uncertainties. The uncertainties can come either from sensor limitations such as occlusions and limited sensor range, or from probabilistic prediction of other road participants, or from unknown social behavior in a new area. To safely and efficiently drive in the presence of these uncertainties, the decision-making and planning modules of autonomous cars should intelligently utilize all available information and appropriately tackle the uncertainties so that proper driving strategies can be generated. In this paper, …
Google Brain（谷歌大脑）；DeepMind（深度思维）；University of Toronto（多伦多大学）；University College London（伦敦大学学院）
To learn directed behaviors in complex environments, intelligent agents need to optimize objective functions. Various objectives are known for designing artificial agents, including task rewards and intrinsic motivation. However, it is unclear how the known objectives relate to each other, which objectives remain yet to be discovered, and which objectives better describe the behavior of humans. We introduce the Action Perception Divergence (APD), an approach for categorizing the space of possible objective functions for embodied agents. We show a spectrum that reaches from narrow to general ob…
The simple gesture of pointing can greatly augment ones ability to comprehend states of the world based on observations. It triggers additional inferences relevant to ones task at hand. We model an agents update to its belief of the world based on individual observations using a partially observable Markov decision process (POMDP), a mainstream artificial intelligence (AI) model of how to act rationally according to beliefs formed through observation. On top of that, we model pointing as a communicative act between agents who have a mutual understanding that the pointed observation must be rel…
Learning Perception to Action Rules through Reinforcement Learning for Theory-of-Mind Reasoning
Saarland University（萨尔兰大学）；Deutsches Forschungszentrum für Künstliche Intelligenz（德国人工智能研究中心）；Télécom Paris（巴黎电信学院）；Max Planck Institute for Informatics（马克斯·普朗克 informatics 研究所）
University of Alberta（阿尔伯塔大学）；University of Oxford（牛津大学）；Technical University Delft（代尔夫特理工大学）
Information gathering in a partially observable environment can be formulated as a reinforcement learning (RL), problem where the reward depends on the agent's uncertainty. For example, the reward can be the negative entropy of the agent's belief over an unknown (or hidden) variable. Typically, the rewards of an RL agent are defined as a function of the state-action pairs and not as a function of the belief of the agent; this hinders the direct application of deep RL methods for such tasks. This paper tackles the challenge of using belief-based rewards for a deep RL agent, by offering a simple…
Belief-Space Control for Personalized Cancer Treatment via Active Inference
American Association for Cancer Research（美国癌症研究协会）；AACR Project GENIE registry（AACR Project GENIE 注册中心）；AACR Project GENIE Biopharma Collaborative（AACR Project GENIE 生物制药合作组织）
University of Amsterdam（阿姆斯特丹大学）；University of Oxford（牛津大学）；University of Liverpool（利物浦大学）；Delft University of Technology（代尔夫特理工大学）
In active perception tasks, an agent aims to select sensory actions that reduce its uncertainty about one or more hidden variables. While partially observable Markov decision processes (POMDPs) provide a natural model for such problems, reward functions that directly penalize uncertainty in the agent's belief can remove the piecewise-linear and convex property of the value function required by most POMDP planners. Furthermore, as the number of sensors available to the agent grows, the computational cost of POMDP planning grows exponentially with it, making POMDP planning infeasible with tradit…
In the classic AGM belief revision theory, beliefs are static and do not change their own shape. For instance, if p is accepted by a rational agent, it will remain p to the agent. But such rarely happens to us. Often, when we accept some information p, what is actually accepted is not the whole p, but only a portion of it; not necessarily because we select the portion but because p must be perceived. Only the perceived p is accepted; and the perception is subject to what we already believe (know). What may, however, happen to the rest of p that initially escaped our attention? In this work we …
HeuDiaSyC-UMR CNRS 6599；Universite de Technologie de Compiegne（贡比涅技术大学）
In current perception systems applied to the rebuilding of the environment for intelligent vehicles, the part reserved to object association for the tracking is increasingly significant. This allows firstly to follow the objects temporal evolution and secondly to increase the reliability of environment perception. We propose in this communication the development of a multi-objects association algorithm with ambiguity removal entering into the design of such a dynamic perception system for intelligent vehicles. This algorithm uses the belief theory and data modelling with fuzzy mathematics in o…
Do SpeechLMs Hear Their Own Opinions? Diagnosing and Mitigating Previous-Belief Contamination in Streaming Emotion Understanding
XJTU-POLIMI Joint School, Xi’an Jiaotong University（西安交通大学-米兰理工大学联合学校（西安交通大学））；University of California, Berkeley（加州大学伯克利分校）；Shenzhen Future Network of Intelligence Institute (FNii-Shenzhen)（深圳未来智能网络研究院）
Active Spiking Perception: The Membrane Potential as a Belief State for Anytime 3D Point Cloud Recognition
Indian Institute of Technology Indore（印度理工学院印多尔分校）；Thadomal Shahani Engineering College, University of Mumbai（孟买大学塔多马尔·沙哈尼工程学院）；Indian Institute of Technology Kharagpur（印度理工学院克勒格布尔分校）；Vellore Institute of Technology Chennai（韦洛尔理工学院金奈分校）；Indiana University Indianapolis（印第安纳大学印第安纳波利斯分校）
Hierarchical Belief Modeling for Zero-Shot Opponent Adaptation in Partially Observable Multi-Agent Navigation
Tsinghua University（清华大学）；Stevens Institute of Technology（史蒂文斯理工学院）；University of California, Los Angeles（加利福尼亚大学洛杉矶分校）；Texas A&M University（德克萨斯A&M大学）
This paper is aimed at providing a uniform framework for reasoning about beliefs of multiple agents and their fusion. In the first part of the paper, we develop logics for reasoning about cautiously merged beliefs of agents with different degrees of reliability. The logics are obtained by combining the multi-agent epistemic logic and multi-sources reasoning systems. Every ordering for the reliability of the agents is represented by a modal operator, so we can reason with the merged results under different situations. The fusion is cautious in the sense that if an agent's belief is in conflict …
British Columbia Institute of Technology（不列颠哥伦比亚理工学院）
Belief revision is the process in which an agent incorporates a new piece of information together with a pre-existing set of beliefs. When the new information comes in the form of a report from another agent, then it is clear that we must first determine whether or not that agent should be trusted. In this paper, we provide a formal approach to modeling trust as a pre-processing step before belief revision. We emphasize that trust is not simply a relation between agents; the trust that one agent has in another is often restricted to a particular domain of expertise. We demonstrate that this fo…
In multi-agent collaboration problems with communication, an agent's ability to encode their intention and interpret other agents' strategies is critical for planning their future actions. This paper introduces a novel algorithm called Intention Embedded Communication (IEC) to mimic an agent's language learning ability. IEC contains a perception module for decoding other agents' intentions in response to their past actions. It also includes a language generation module for learning implicit grammar during communication with two or more agents. Such grammar, by construction, should be compact f…
Université Paris 1 Panthéon-Sorbonne（巴黎第一大学）；Quarkslab SA（Quarkslab公司）；Université Paris-Dauphine, PSL University（巴黎第九大学）
In this paper, we address the problem of finding a correspondence, or matching, between the functions of two programs in binary form, which is one of the most common task in binary diffing. We introduce a new formulation of this problem as a particular instance of a graph edit problem over the call graphs of the programs. In this formulation, the quality of a mapping is evaluated simultaneously with respect to both function content and call graph similarities. We show that this formulation is equivalent to a network alignment problem. We propose a solving strategy for this problem based on max…
Working together on complex collaborative tasks requires agents to coordinate their actions. Doing this explicitly or completely prior to the actual interaction is not always possible nor sufficient. Agents also need to continuously understand the current actions of others and quickly adapt their own behavior appropriately. Here we investigate how efficient, automatic coordination processes at the level of mental states (intentions, goals), which we call belief resonance, can lead to collaborative situated problem-solving. We present a model of hierarchical active inference for collaborative a…
