Why generalization in rl is difficult: Epistemic pomdps and implicit partial observability
Generalization is a central challenge for the deployment of reinforcement learning (RL)
systems in the real world. In this paper, we show that the sequential structure of the RL …
systems in the real world. In this paper, we show that the sequential structure of the RL …
F2a2: Flexible fully-decentralized approximate actor-critic for cooperative multi-agent reinforcement learning
Traditional centralized multi-agent reinforcement learning (MARL) algorithms are sometimes
unpractical in complicated applications due to non-interactivity between agents, the curse of …
unpractical in complicated applications due to non-interactivity between agents, the curse of …
The geometry of memoryless stochastic policy optimization in infinite-horizon POMDPs
J Müller, G Montúfar - arXiv preprint arXiv:2110.07409, 2021 - arxiv.org
We consider the problem of finding the best memoryless stochastic policy for an infinite-
horizon partially observable Markov decision process (POMDP) with finite state and action …
horizon partially observable Markov decision process (POMDP) with finite state and action …
Policy gradient in partially observable environments: Approximation and convergence
Policy gradient is a generic and flexible reinforcement learning approach that generally
enjoys simplicity in analysis, implementation, and deployment. In the last few decades, this …
enjoys simplicity in analysis, implementation, and deployment. In the last few decades, this …
Active Inference and Reinforcement Learning: A Unified Inference on Continuous State and Action Spaces under Partial Observability
P Malekzadeh, KN Plataniotis - Neural Computation, 2024 - direct.mit.edu
Reinforcement learning (RL) has garnered significant attention for developing decision-
making agents that aim to maximize rewards, specified by an external supervisor, within fully …
making agents that aim to maximize rewards, specified by an external supervisor, within fully …
Geometry of Optimization in Markov Decision Processes and Neural Network-Based PDE Solvers
J Müller - 2023 - ul.qucosa.de
Abstract (EN) This thesis is divided into two parts dealing with the optimization problems in
Markov decision processes (MDPs) and different neural network-based numerical solvers …
Markov decision processes (MDPs) and different neural network-based numerical solvers …
Open problem: Approximate planning of pomdps in the class of memoryless policies
K Azizzadenesheli, A Lazaric… - … on Learning Theory, 2016 - proceedings.mlr.press
Planning plays an important role in the broad class of decision theory. Planning has drawn
much attention in recent work in the robotics and sequential decision making areas …
much attention in recent work in the robotics and sequential decision making areas …
Algebraic optimization of sequential decision problems
We study the optimization of the expected long-term reward in finite partially observable
Markov decision processes over the set of stationary stochastic policies. In the case of …
Markov decision processes over the set of stationary stochastic policies. In the case of …
Learning Algorithms for Intelligent Agents and Mechanisms
J Rahme - 2022 - search.proquest.com
The ability to learn from past experiences and adapt one's behavior accordingly within an
environment or context to achieve a certain goal is a characteristic of a truly intelligent entity …
environment or context to achieve a certain goal is a characteristic of a truly intelligent entity …
Combining information-seeking exploration and reward maximization: Unified inference on continuous state and action spaces under partial observability
P Malekzadeh, KN Plataniotis - arXiv preprint arXiv:2212.07946, 2022 - arxiv.org
Reinforcement learning (RL) gained considerable attention by creating decision-making
agents that maximize rewards received from fully observable environments. However, many …
agents that maximize rewards received from fully observable environments. However, many …