Filters








5,984 Hits in 7.8 sec

Simultaneous Perception-Action Design via Invariant Finite Belief Sets [article]

Michael Hibbard, Takashi Tanaka, Ufuk Topcu
2021 arXiv   pre-print
To alleviate these perception costs, we develop a novel simultaneous perception-action design framework wherein an agent senses only the task-relevant information.  ...  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  ...  Simultaneous Perception-Action Design We now formulate the simultaneous perception-action design (SPADE) problem of the agent.  ... 
arXiv:2109.05073v1 fatcat:x7kdceuafnfuffgv23ugjuidqq

Computing Environment-Aware Agent Behaviours with Logic Program Updates [chapter]

José J. Alferes, Antonio Brogi, João A. Leite, Luís M. Pereira
2002 Lecture Notes in Computer Science  
Logic program updates are specified, in our setting, by the language of updates LUPS.  ...  The obtained declarative semantics sets a firm basis for the development, implementation, and analysis of environmentaware agents.  ...  sets of perceptions of the environment that P may (simultaneously) get.  ... 
doi:10.1007/3-540-45607-4_12 fatcat:z3hrk22me5gvrm3dudi4lxuhbe

What does an artificial design agent mean by being 'situated'?

Gregory J. Smith, John S. Gero
2005 Design Studies  
Both the problem and solution of many designing tasks emerge through this "conversation" between a situated designer and the medium of the design.  ...  Schön described designing as a "conversation with materials conducted in the medium of drawing".  ...  The agent can learn by interacting with the environment, performing actions and receiving percepts via its own sensors, comparing predictions of existing models against received percepts, and adapting  ... 
doi:10.1016/j.destud.2005.01.001 fatcat:uii6eb6dtvefzdd6777kar3d4q

Communication, Risk, and Efficiency in Games

Andreas Blume
1998 Games and Economic Behavior  
The paper identifies stable outcomes under population learning dynamics in which individuals with some probability replace their current strategy with a best reply against beliefs supported on a sample  ...  However, for Γ 2 ; the set Q is not invariant under the dynamic process.  ...  In the words of CDFR: "This doubt about the action of a receiver is overcome by the two-way communication design since both players receive information about the likely play of their opponents.  ... 
doi:10.1006/game.1997.0598 fatcat:5s5x2tlftzgw5g7ycrafbxaia4

Hybrid Verification Technique for Decision-Making of Self-Driving Vehicles

Mohammed Al-Nuaimi, Sapto Wibowo, Hongyang Qu, Jonathan Aitken, Sandor Veres
2021 Journal of Sensor and Actuator Networks  
The overall structure of our autonomous vehicle (AV) system consists of: (1) A perception system of sensors that feeds data into (2) a rational agent (RA) based on a belief–desire–intention (BDI) architecture  ...  MCMAS is used to check the consistency and stability of the BDI agent logic during design-time.  ...  Pre-programmed rules are used to set the relationship between the perception predicates (beliefs) and the available actions.  ... 
doi:10.3390/jsan10030042 fatcat:vbui7h3oqbbc5fxqubkcf44534

A Survey on Active Simultaneous Localization and Mapping: State of the Art and New Frontiers [article]

Julio A. Placed, Jared Strader, Henry Carrillo, Nikolay Atanasov, Vadim Indelman, Luca Carlone, José A. Castellanos
2022 arXiv   pre-print
Active Simultaneous Localization and Mapping (SLAM) is the problem of planning and controlling the motion of a robot to build the most accurate and complete model of the surrounding environment.  ...  We then analyze alternative approaches, including belief-space planning and modern techniques based on deep reinforcement learning, and review related work on multi-robot coordination.  ...  On the Reward Function Design and the Action Set Tai and Liu [188] are among the first to employ DRL for robotic exploration, extracting the next best actions to execute from a discrete set using a 2  ... 
arXiv:2207.00254v1 fatcat:ksyanwtftje3xkx7tbql6hry54

Intelligent Agents: The Key Concepts [chapter]

Michael Wooldridge
2002 Lecture Notes in Computer Science  
Second, reactive architectures are discussed, in which symbolic representations and models are eschewed in favour of a closer relationship between agent perception and action.  ...  and finally to intentions and actions.  ...  An environment is discrete if there are a fixed, finite number of actions and percepts in it.  ... 
doi:10.1007/3-540-45982-0_1 fatcat:fiazijq3yjacvnike7poaqtha4

Planning and Decision-Making for Autonomous Vehicles

Wilko Schwarting, Javier Alonso-Mora, Daniela Rus
2018 Annual Review of Control Robotics and Autonomous Systems  
In this survey, we emphasize recent approaches for integrated perception and planning and for behavior-aware planning, many of which rely on machine learning.  ...  Recent advances in the field of perception, planning, and decision-making for autonomous vehicles have led to great improvements in functional capabilities, with several prototypes already driving on our  ...  Classical perception systems extract information in the form of manually designed features from raw sensory data.  ... 
doi:10.1146/annurev-control-060117-105157 fatcat:hgrhw76idbbdrct742bbhnsqem

Estimating Rationally Inattentive Utility Functions with Deep Clustering for Framing - Applications in YouTube Engagement Dynamics [article]

William Hoiles, Vikram Krishnamurthy
2018 arXiv   pre-print
The set of possible signals S(α) for a given attention strategy α is finite.  ...  Constructing Preference and Policy Invariant Frames via Deep Learning Here a deep embedding method is provided that learns the policy invariant frames of the agent.  ... 
arXiv:1812.09640v1 fatcat:uwg4z4qrpbfndpt2xmewbgpuwy

Reclaiming saliency: rhythmic precision-modulated action and perception [article]

Ajith Anil Meera, Filip Novicky, Thomas Parr, Karl Friston, Pablo Lanillos, Noor Sajid
2022 arXiv   pre-print
However, they fail to consider the circular causality between perception and action. In other words, they do not consider where to sample next, given current beliefs.  ...  Briefly, we associate attention with precision control, i.e., the confidence with which beliefs can be updated given sampled sensory data, and salience with uncertainty minimisation that underwrites the  ...  The precision-modulated perception is formalised for a robotics setting; via (i) state estimation (i.e., estimating the hidden states of a dynamic system from sensory signals -Sec. 4.2.2), and (ii) system  ... 
arXiv:2203.12652v1 fatcat:mmc3cqzjlbdylolhj2m6qq4thq

Evolution of Players' Misperceptions in Hypergames Under Perfect Observations

Bahman Gharesifard, Jorge Cortes
2012 IEEE Transactions on Automatic Control  
The focus is on developing methods to modify the players' perception about other players' preferences by incorporating the lessons learned from observing their actions.  ...  We introduce the novel notion of H-digraph as a useful tool to encode the information in a hypergame, and fully characterize how this digraph is affected by changes in the players' beliefs.  ...  Basic notions A 0-level hypergame is a finite game, i.e., a triplet G = (V, S otcm , P), where V is a set of n players; S otcm = S 1 × . . . × S n is the outcome set, where S i is a finite set of strategies  ... 
doi:10.1109/tac.2011.2180113 fatcat:tl3h3fgc3zfjjhlje2mqv3gspy

Sorting Objects from a Conveyor Belt Using POMDPs with Multiple-Object Observations and Information-Gain Rewards

Ady-Daniel Mezei, Levente Tamás, Lucian Buşoniu
2020 Sensors  
The DESPOT algorithm is applied to solve the problem, automatically obtaining a sequence of observation viewpoints and class decision actions.  ...  describe this task in the framework of partially observable Markov decision processes, and we propose a reward function that explicitly takes into account the information gain of the viewpoint selection actions  ...  • A is a set of actions available to the robot, again discrete and finite. Actions are a ∈ A.  ... 
doi:10.3390/s20092481 pmid:32349393 pmcid:PMC7249059 fatcat:pnglxuicobgmhn566vatmj376y

Learning About a Class of Belief-Dependent Preferences Without Information on Beliefs

Charles Bellemare, Alexander Sebald
2011 Social Science Research Network  
We show how to bound the effect of belief-dependent preferences on choices in sequential two-player games without information about the (higher-order) beliefs of players.  ...  We show how the size of the bounds can be substantially reduced by exploiting a specific invariance property common to preferences in this class.  ...  At each non-terminal history h ∈ H\Z each player i ∈ N has a nonempty, finite set of feasible actions A i,h . A typical element of A i,h is denoted by a i,h .  ... 
doi:10.2139/ssrn.1923271 fatcat:mm2jfr3s4vauhovt7dftkcif4q

Game Theory, Conditional Preferences, and Social Influence

Wynn C. Stirling, Teppo Felin, Attila Szolnoki
2013 PLoS ONE  
Our model is generalizable to these settings and might thus also be applied to settings such as the design of artificial or expert systems, the aggregation of information, or coordination within, or design  ...  ,X n g, n §2, denote a set of players, and let A i denote a finite space of feasible actions from which X i may choose one element. A profile is an array a~(a 1 , . . . ,a n )[A 1 | Á Á Á |A n .  ... 
doi:10.1371/journal.pone.0056751 pmid:23451078 pmcid:PMC3581537 fatcat:34jzu2vakvaflh6rocs5fpwpjq

Elements of Consciousness and Cognition. Biology, Mathematic, Physics and Panpsychism: an Information Topology Perspective [article]

Pierre Baudot
2018 arXiv   pre-print
It underlines the diversity of the learning mechanisms that sustain the dynamics of perception and consciousness, including adaptive and homeostatic processes on multiple scales.  ...  The first chapter presents the postulates and results on elementary perception at various organizational scales of the nervous system and proposes the hypothesis of an electrodynamic intrinsic nature of  ...  Dynamics, Geometries, action invariance and Homeostasis of consciousness The invariances to group actions of perception "The research of invariant is the fundamental fact of perception"; this is in essence  ... 
arXiv:1807.04520v2 fatcat:f2qyihx7cnadjl2ctkqn2f3j2q
« Previous Showing results 1 — 15 out of 5,984 results