Human-level reinforcement learning performance of recurrent neural networks is linked to hyperperseveration, not directed exploration
A key feature of animal and human decision-making is to balance exploring unknown
options for information gain (directed exploration) versus exploiting known options for …
options for information gain (directed exploration) versus exploiting known options for …
[HTML][HTML] Exploration–Exploitation Mechanisms in Recurrent Neural Networks and Human Learners in Restless Bandit Problems
A key feature of animal and human decision-making is to balance the exploration of
unknown options for information gain (directed exploration) versus selecting known options …
unknown options for information gain (directed exploration) versus selecting known options …
Impact of multi-armed bandit strategies on deep recurrent reinforcement learning
V Zangirolami, M Borrotti - arXiv preprint arXiv:2310.08331, 2023 - arxiv.org
Incomplete knowledge of the environment leads an agent to make decisions under
uncertainty. One of the major dilemmas in Reinforcement Learning (RL) where an …
uncertainty. One of the major dilemmas in Reinforcement Learning (RL) where an …
Distilling human decision-making dynamics: a comparative analysis of low-dimensional architectures
Recent advances in examining biological decision-making behaviors have increasingly
favored recurrent neural networks (RNNs) over traditional cognitive models grounded in …
favored recurrent neural networks (RNNs) over traditional cognitive models grounded in …
[HTML][HTML] Dealing with uncertainty: Balancing exploration and exploitation in deep recurrent reinforcement learning
V Zangirolami, M Borrotti - Knowledge-Based Systems, 2024 - Elsevier
Incomplete knowledge of the environment leads an agent to make decisions under
uncertainty. One of the major dilemmas in Reinforcement Learning (RL) where an …
uncertainty. One of the major dilemmas in Reinforcement Learning (RL) where an …
A meta reinforcement learning account of behavioral adaptation to volatility in recurrent neural networks
Natural environments often exhibit various degrees of volatility, ranging from slowly
changing to rapidly changing contingencies. How learners adapt to changing environments …
changing to rapidly changing contingencies. How learners adapt to changing environments …
Reinforcement learning with fast and forgetful memory
Nearly all real world tasks are inherently partially observable, necessitating the use of
memory in Reinforcement Learning (RL). Most model-free approaches summarize the …
memory in Reinforcement Learning (RL). Most model-free approaches summarize the …
Connecting exploration, generalization, and planning in correlated trees
Human reinforcement learning (RL) is characterized by different challenges. Exploration has
been studied extensively in multi-armed bandits, while planning has been investigated in …
been studied extensively in multi-armed bandits, while planning has been investigated in …
[HTML][HTML] Harnessing the flexibility of neural networks to predict dynamic theoretical parameters underlying human choice behavior
Reinforcement learning (RL) models are used extensively to study human behavior. These
rely on normative models of behavior and stress interpretability over predictive capabilities …
rely on normative models of behavior and stress interpretability over predictive capabilities …
Top-down signaling dynamically mediates information processing in biologically inspired RNNs
Recent studies have proposed employing biologically plausible recurrent neural networks
(RNNs) to explore flexible decision making processes in the brain. However, the …
(RNNs) to explore flexible decision making processes in the brain. However, the …