78,512 Hits in 5.6 sec

Using Logical Specifications of Objectives in Multi-Objective Reinforcement Learning [article]

Kolby Nottingham, Anand Balakrishnan, Jyotirmoy Deshmukh, David Wingate
2021 arXiv   pre-print
The multi-objective reinforcement learning (MORL) framework separates a reward function into several objectives.  ...  We propose using propositional logic to specify the importance of multiple objectives.  ...  Multi-Objective Reinforcement Learning We alter the traditional RL formulation as an MDP by using a vector of rewards rather than a scalar reward, resulting in a Multi-Objective Markov Decision Process  ... 
arXiv:1910.01723v3 fatcat:n77zuwbivzb3jgzpkspksmsnum

A Survey on Interpretable Reinforcement Learning [article]

Claire Glanois, Paul Weng, Matthieu Zimmer, Dong Li, Tianpei Yang, Jianye Hao, Wulong Liu
2021 arXiv   pre-print
This survey provides an overview of various approaches to achieve higher interpretability in reinforcement learning (RL).  ...  Although deep reinforcement learning has become a promising machine learning approach for sequential decision-making problems, it is still not mature enough for high-stake domains such as autonomous driving  ...  Zhang, “Object-Oriented Dynamics Learning through Multi-Level Abstraction,” in AAAI, 2020. [138] W. Agnew and P.  ... 
arXiv:2112.13112v1 fatcat:emsedwo3xfc5pmvjexujnqaone

Review, Analyze, and Design a Comprehensive Deep Reinforcement Learning Framework [article]

Ngoc Duy Nguyen, Thanh Thi Nguyen, Hai Nguyen, Saeid Nahavandi
2020 arXiv   pre-print
However, current research interests are diverted into different directions, such as multi-agent and multi-objective learning, and human-machine interactions.  ...  Finally, to enforce generalization, the proposed architecture does not depend on a specific RL algorithm, a network configuration, the number of agents, or the type of agents.  ...  Multi-objective environment and Use a multi-objective learner (MO Q-Learning) fruit/samples/basic/multi objectives multi-objective RL to train an agent to play Mountain Car [47] 5.  ... 
arXiv:2002.11883v1 fatcat:yziq6kwryvh5hiwjm6ju2r5srq

Qualitative comparison of graph-based and logic-based multi-relational data mining

Nikhil S. Ketkar, Lawrence B. Holder, Diane J. Cook
2005 Proceedings of the 4th international workshop on Multi-relational mining - MRDM '05  
The goal of this paper is to generate insights about the differences between graph-based and logic-based approaches to multi-relational data mining by performing a case study of graph-based system, Subdue  ...  It is also observed that CProgol is better at learning semantically complicated concepts and it tends to use background knowledge more effectively than Subdue.  ...  The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of AFRL or the  ... 
doi:10.1145/1090193.1090198 fatcat:eyj43infanbvzndaeozwngh4ce

Introduction to the special issue on learning semantics

Antoine Bordes, Léon Bottou, Ronan Collobert, Dan Roth, Jason Weston, Luke Zettlemoyer
2013 Machine Learning  
A growing number of efforts to develop machine learning approaches for semantic analysis now aim to find (in an automated way) these interpretations (Miller et al.  ...  Computational linguists use the term semantics (Lewis 1970) to refer to the possible interpretations of natural language sentences.  ...  This special issue would not have been possible without the contributions of many people. We wish to sincerely thank all the authors for submitting their work to this special issue.  ... 
doi:10.1007/s10994-013-5381-4 fatcat:7i5ubznmabewldc5asxj3xqfru

Autonomous driving: cognitive construction and situation understanding

Shitao Chen, Zhiqiang Jian, Yuhao Huang, Yu Chen, Zhuoli Zhou, Nanning Zheng
2019 Science China Information Sciences  
It further describes how to use multi-sensor and graph data with semantic information (such as traffic maps and a spatial correlation of events) to realize the associative representations of objects and  ...  In current research of autonomous driving, the most widely adopted technique is to use a basic framework of serial information processing and computations, which consists of four modules: perception, planning  ...  indoor robot mapping and navigation through deep reinforcement learning, showing the adaptability of deep reinforcement learning in semi-open scenes.  ... 
doi:10.1007/s11432-018-9850-9 fatcat:qys3uucz3zgznfou6vgfjerwlq

Reinforcement Learning with Probabilistic Guarantees for Autonomous Driving [article]

Maxime Bouton, Jesper Karlsson, Alireza Nakhaei, Kikuo Fujimura, Mykel J. Kochenderfer, Jana Tumova
2019 arXiv   pre-print
Reinforcement learning (RL) has been used to automatically derive suitable behavior in uncertain environments, but it does not provide any guarantee on the performance of the resulting policy.  ...  An exploration strategy is derived prior to training that constrains the agent to choose among actions that satisfy a desired probabilistic specification expressed with linear temporal logic (LTL).  ...  The reward function used in reinforcement learning might express a different objective than the one specified through the LTL formula.  ... 
arXiv:1904.07189v2 fatcat:fxsjjitdeven3a56jn6oxxohbm

A Hybrid Algorithm in Reinforcement Learning for Crowd Simulation

2020 International journal of recent technology and engineering  
Two kinds of applications are used in Reinforcement Learning such as tracking applications and transportation monitoring applications for pretending the crowd sizes.  ...  The proposed Hybrid Agent Reinforcement Learning (HARL) algorithm combines the Q-Learning off-policy value function and SARSA algorithm on-policy value function, which is used for dynamic crowd evacuation  ...  For the multi-object tracking a CRL (Collaborative Reinforcement Learning) method is intended to propose in this paper.  ... 
doi:10.35940/ijrte.f9187.038620 fatcat:3a65mumhdzfchdtqvy4vrwkzku

Neural Logic Machines [article]

Honghua Dong, Jiayuan Mao, Tian Lin, Chong Wang, Lihong Li, Denny Zhou
2019 arXiv   pre-print
NLMs exploit the power of both neural networks---as function approximators, and logic programming---as a symbolic processor for objects with properties, relations, logic connectives, and quantifiers.  ...  We propose the Neural Logic Machine (NLM), a neural-symbolic architecture for both inductive learning and logic reasoning.  ...  In the experiments, Softmax-Cross-Entropy loss is used for supervised learning tasks, and REINFORCE (Williams, 1992) is used for reinforcement learning tasks.  ... 
arXiv:1904.11694v1 fatcat:6umfj3sbubcgfiputfxf6urn5i

Multi-Agent Reinforcement Learning with Temporal Logic Specifications [article]

Lewis Hammond and Alessandro Abate and Julian Gutierrez and Michael Wooldridge
2021 arXiv   pre-print
In order to overcome this limitation, we develop the first multi-agent reinforcement learning technique for temporal logic specifications, which is also novel in its ability to handle multiple specifications  ...  In this paper, we study the problem of learning to satisfy temporal logic specifications with a group of agents in an unknown environment, which may exhibit probabilistic behaviour.  ...  Hammond acknowledges the support of an EPSRC Doctoral Training Partnership studentship (Reference: 2218880) and the University of Oxford ARC facility. 2 Wooldridge and Abate acknowledge the support of  ... 
arXiv:2102.00582v2 fatcat:wq2vons5sbhkbjjvq3iykksocy

The segregation of vocal circuits solves a credit assignment problem associated with multi-objective reinforcement learning [article]

Don Murdoch, Ruidong Chen, Jesse H Goldberg
2017 bioRxiv   pre-print
Non-global, target-specific reinforcement signals have established utility in machine implementation of multi-objective learning.  ...  Strobe light negatively reinforced place learning but did not affect song syllable learning. Noise bursts positively reinforced place preference but negatively reinforced syllable learning.  ...  Non-global, target-specific reinforcement signals have established utility in machine implementation of multi-objective learning.  ... 
doi:10.1101/236273 fatcat:uvyzmswrb5cafnjvlnppvbg7ji

Flexible and Efficient Long-Range Planning Through Curious Exploration [article]

Aidan Curtis, Minjian Xin, Dilip Arumugam, Kevin Feigelis, Daniel Yamins
2020 arXiv   pre-print
In contrast, deep reinforcement learning (DRL) methods use flexible neural-network-based function approximators to discover policies that generalize naturally to unseen circumstances.  ...  Identifying algorithms that flexibly and efficiently discover temporally-extended multi-phase plans is an essential step for the advancement of robotics and model-based reinforcement learning.  ...  Grandmaster level in starcraft ii using multi-agent reinforcement learning. Nature, pp. 1-5, 2019.Wang, Z., Garrett, C. R., Kaelbling, L. P., and Lozano-Perez, T.  ... 
arXiv:2004.10876v2 fatcat:4ioyp3qubnhvtg7txjjkzuixp4

Hypervolume-Based Multi-Objective Reinforcement Learning [chapter]

Kristof Van Moffaert, Madalina M. Drugan, Ann Nowé
2013 Lecture Notes in Computer Science  
In this paper, we propose a novel on-line multi-objective reinforcement learning (MORL) algorithm that uses the hypervolume indicator as an action selection strategy.  ...  In reinforcement learning (RL), introducing a quality indicator in an algorithm's decision logic was not attempted before.  ...  In Section 2, we provide an overview of background concepts such as multi-objective optimization and we introduce reinforcement learning in Section 3.  ... 
doi:10.1007/978-3-642-37140-0_28 fatcat:k2zbgno66rdtzdq7yob7quzp5y

Conditional Learning of Rules and Plans by Knowledge Exchange in Logical Agents [chapter]

Stefania Costantini, Pierangelo Dell'Acqua, Luís Moniz Pereira
2011 Lecture Notes in Computer Science  
In this paper we introduce a form of cooperation among agents based on exchanging sets of rules.  ...  In principle, the approach extends to agent societies a feature which is proper of human societies, i.e., the cultural transmission of abilities.  ...  In a multi-agent setting however, other forms of learning can be introduced that, though related to the classical ones, are specifically tailored to multi-agent systems (MAS) topics and issues.  ... 
doi:10.1007/978-3-642-22546-8_20 fatcat:ekzlg46mgvbmnppcfn76lkw2ii

A novel adaptive weight selection algorithm for multi-objective multi-agent reinforcement learning

Kristof Van Moffaert, Tim Brys, Arjun Chandra, Lukas Esterle, Peter R. Lewis, Ann Nowe
2014 2014 International Joint Conference on Neural Networks (IJCNN)  
Our first contribution is a thorough analysis of well known scalarization schemes within the multi-objective multi-agent reinforcement learning setup.  ...  While the field of multi-objective optimization has made many advances in applying scalarization techniques to obtain good solution trade-offs, the utility of applying these techniques in the multi-objective  ...  Multi-objective reinforcement learning Multi-objective reinforcement learning (MORL) is an extension to standard reinforcement learning where the environment consists of two or more feedback signals, i.e  ... 
doi:10.1109/ijcnn.2014.6889637 dblp:conf/ijcnn/MoffaertBCELN14 fatcat:cdg6m4gzo5amxeswcg6egcahue
« Previous Showing results 1 — 15 out of 78,512 results