Partially observable Markov decision processes and performance sensitivity analysis
PulseAugur coverage of Partially observable Markov decision processes and performance sensitivity analysis — every cluster mentioning Partially observable Markov decision processes and performance sensitivity analysis across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
New algorithms enable quasipolynomial-time planning for decodable POMDPs
Researchers have developed new algorithms for planning in partially observable Markov decision processes (POMDPs). These algorithms achieve quasipolynomial-time approximate planning for a class of POMDPs known as $\rho$…
-
New active inference method enables efficient model learning for planning
Researchers have developed a new method for parameter learning in active inference, which frames perception, learning, and planning as variational inference. This approach uses an augmented generative model with epistem…
-
New research details complexity gap between POMDPs and multi-environment POMDPs
A new research paper published on arXiv explores the computational complexity of Partially Observable Markov Decision Processes (POMDPs) and their multi-environment counterparts (ME-POMDPs). The study identifies specifi…
-
Frozen recurrent networks offer efficient baseline for partially observable RL
A new research paper explores the necessity of learning recurrent networks in reinforcement learning (RL) for partially observable environments. The study proposes using a frozen recurrent network combined with a traina…
-
FLoRa architecture optimizes UAV data collection from energy-constrained IoT devices
Researchers have developed FLoRa, a novel architecture for data collection from energy-constrained LoRa IoT devices using Unmanned Aerial Vehicles (UAVs). FLoRa employs a multi-level optimization approach, integrating S…
-
New framework enhances robotic planning in complex environments
Researchers have developed a new framework for online planning in robotic systems operating in complex, partially observable environments. This approach, an extension of the Rao-Blackwellized online POMDP (RB-POMDP) fra…
-
New framework unifies uncertainty reduction and reward in POMDPs
Researchers have developed a new framework for Partially Observable Markov Decision Processes (POMDPs) that directly rewards uncertainty reduction. This approach, termed minimizing Expected Free Energy (EFE), is shown t…
-
New PO-PDDL formulation enables LLM-friendly robot planning under uncertainty
Researchers have developed PO-PDDL, a new symbolic formulation for Partially Observable Markov Decision Processes (POMDPs) that maintains the relational structure of PDDL while explicitly modeling partial observability …
-
ReAct Pattern Enhances LLM Reasoning and Action Capabilities
The ReAct Pattern is a design pattern for Large Language Models (LLMs) that enhances their reasoning and action capabilities in complex environments. It enables LLMs to perceive, reason, and act, allowing them to learn …
-
POMDP value functions characterized as semi-algebraic sets
Researchers have characterized the feasible set of value functions in partially observable Markov decision processes (POMDPs) as a semi-algebraic set. This extends previous work on fully observable processes, revealing …
-
New framework synthesizes POMDP policies using sampling and model-checking
Researchers have developed a new framework to synthesize policies for Partially Observable Markov Decision Processes (POMDPs), which are used for decision-making under uncertainty. This approach combines sampling-based …
-
New MCTS analysis offers theoretical guarantees for POMDP planning
Researchers have developed a new finite-time analysis for Monte Carlo Tree Search (MCTS) when applied to Partially Observable Markov Decision Processes (POMDPs). This work provides probabilistic concentration bounds for…