Audio note: this article contains 363 uses of latex notation, so the narration may be difficult to follow. There's a link to the original text in the episode description.

Introduction

In this post, we describe a generalization of Markov decision processes (MDPs) and partially observable Markov decision processes (POMDPs) called crisp supra-MDPs and supra-POMDPs. The new feature of these decision processes is that the stochastic transition dynamics are multivalued, i.e. specified by credal sets. We describe how supra-MDPs give rise to crisp causal laws, the hypotheses of infra-Bayesian reinforcement learning. Furthermore, we discuss how supra-MDPs can approximate MDPs by a coarsening of the state space. This coarsening allows an agent to be agnostic about the detailed dynamics while still having performance guarantees for the full MDP.

Analogously to the classical theory, we describe an algorithm to compute a Markov optimal policy for supra-MDPs with finite time [...]

---

Outline:

(00:22) Introduction

(01:41) Supra-Markov Decision Processes

(01:45) Defining supra-MDPs

(06:55) Supra-bandits

(08:02) Crisp causal laws arising from supra-MDPs and (supra-)RDPs

(19:25) Approximating an MDP with a supra-MDP

(22:42) Computing the Markov optimal policy for finite time horizons

(25:02) Existence of stationary optimal policy

(28:18) Interpreting supra-MDPs as stochastic games

(32:40) Regret bounds for episodic supra-MDPs

(34:16) Supra-Partially Observable Markov Decision Processes

(36:16) Equivalence of crisp causal laws and crisp supra-POMDPs

(37:18) Proof of Proposition 2

(41:21) Acknowledgements

The original text contained 8 footnotes which were omitted from this narration.

---

First published:

Source:

---

Narrated by TYPE III AUDIO.

---

Images from the article:

Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.