Markov Models
Learning in Non-Cooperative Configurable Markov Decision Processes Giorgia Ramponi ETH AI Center Zurich, Switzerland gramponi@ethz.ch Alberto Maria Metelli Politecnico di Milano Milan, Italy
Reinforcement Learning agent and a configurator that can modify some environmental parameters to improve the agent's performance.
Average-Reward Learning and Planning with Options Yi Wan, Abhishek Naik, Richard S. Sutton {wan6,anaik1,rsutton }@ualberta.ca University of Alberta, Amii
We extend the options framework for temporal abstraction in reinforcement learning from discounted Markov decision processes (MDPs) to average-reward MDPs. Our contributions include general convergent off-policy inter-option learning algorithms, intra-option algorithms for learning values and models, as well as sample-based planning variants of our learning algorithms. Our algorithms and convergence proofs extend those recently developed by Wan, Naik, and Sutton.
Optimal Scaling for Locally Balanced Proposals in Discrete Spaces
The Markov Chain Monte Carlo (MCMC) algorithm is one of the most widely used methods for sampling from intractable distributions (Robert & Casella, 2013).
Distinguishing discrete and continuous behavioral variability using warped autoregressive HMMs
W ARHMM achieves similar performance to the standard ARHMM while using fewer behavioral syllables.
First-Explore, then Exploit: Meta-Learning to Solve Hard Exploration-Exploitation Trade-Offs Ben Norman
The objective is to maximize the total reward accumulated over all episodes (e.g., the number of games won), expressed as
Towards Minimax Optimal Reinforcement Learning in Factored Markov Decision Processes
We study minimax optimal reinforcement learning in episodic factored Markov decision processes (FMDPs), which are MDPs with conditionally independent transition components.
Experimental design for MRI by greedy policy search
In today's clinical practice, magnetic resonance imaging (MRI) is routinely accelerated through subsampling of the associated Fourier domain.
We thank the reviewers for their time and thorough comments, as well as their valuation of our work including its
For the larger discussion items, please find the detailed comments below. Additionally, the reviewers highlighted the importance of quantitative fits. We currently attempt to differentiate between these models using additional manipulations. R-learning may be advantageous for computation. Our work builds upon results in the field including Ref [2] This observation enabled us to pursue the hypothesis of the leaky estimate of average reward.