Sparse Multi-task Reinforcement Learning - Archive ouverte HAL Access content directly
Conference Papers Year : 2014

Daniele Calandriello
• Function : Author
• PersonId : 960706
Alessandro Lazaric
Marcello Restelli
• Function : Author
• PersonId : 960707

#### Abstract

In multi-task reinforcement learning (MTRL), the objective is to simultaneously learn multiple tasks and exploit their similarity to improve the performance w.r.t.\ single-task learning. In this paper we investigate the case when all the tasks can be accurately represented in a linear approximation space using the same small subset of the original (large) set of features. This is equivalent to assuming that the weight vectors of the task value functions are \textit{jointly sparse}, i.e., the set of their non-zero components is small and it is shared across tasks. Building on existing results in multi-task regression, we develop two multi-task extensions of the fitted $Q$-iteration algorithm. While the first algorithm assumes that the tasks are jointly sparse in the given representation, the second one learns a transformation of the features in the attempt of finding a more sparse representation. For both algorithms we provide a sample complexity analysis and numerical simulations.

#### Domains

Statistics [stat] Machine Learning [stat.ML]

### Dates and versions

hal-01073513 , version 1 (31-01-2015)

### Identifiers

• HAL Id : hal-01073513 , version 1

### Cite

Daniele Calandriello, Alessandro Lazaric, Marcello Restelli. Sparse Multi-task Reinforcement Learning. NIPS - Advances in Neural Information Processing Systems 26, Dec 2014, Montreal, Canada. ⟨hal-01073513⟩

### Export

BibTeX TEI Dublin Core DC Terms EndNote Datacite

224 View