#### Filter Results:

- Full text PDF available (34)

#### Publication Year

2001

2018

- This year (1)
- Last 5 years (16)
- Last 10 years (40)

#### Publication Type

#### Co-author

#### Journals and Conferences

Learn More

- Tor Lattimore, Marcus Hutter, Peter Sunehag
- ICML
- 2013

We present a new algorithm for general reinforcement learning where the true environment is known to belong to a finite class of N arbitrary models. The algorithm is shown to be near-optimal for allâ€¦ (More)

- O. Thomas, Peter Sunehag, +6 authors P. Saunders
- Pervasive and Mobile Computing
- 2010

Detailed monitoring of training sessions of elite athletes is an important component of their training. In this paper we describe an application that performs a precise segmentation and labeling ofâ€¦ (More)

- Matthew W. Robards, Peter Sunehag
- 2009 Ninth IEEE International Conference on Dataâ€¦
- 2009

Subsequence clustering aims to find patterns that appear repeatedly in time series data. We introduce a novel subsequence clustering technique that we call semi-Markov kmeans clustering. Theâ€¦ (More)

- Gabriel Dulac-Arnold, Richard Evans, +7 authors Ben Coppin
- 2015

Being able to reason in an environment with a large number of discrete actions is essential to bringing reinforcement learning to a larger class of problems. Recommender systems, industrial plantsâ€¦ (More)

- Peter Sunehag, Jochen Trumpf, S. V. N. Vishwanathan, Nicol N. Schraudolph
- AISTATS
- 2009

We provide a variable metric stochastic approximation theory. In doing so, we provide a convergence theory for a large class of online variable metric methods including the recently introduced onlineâ€¦ (More)

- Peter Sunehag, Marcus Hutter
- ALT
- 2010

We are studying long term sequence prediction (forecasting). We approach this by investigating criteria for choosing a compact useful state representation. The state is supposed to summarize usefulâ€¦ (More)

- Peter Sunehag, Marcus Hutter
- Australasian Conference on Artificialâ€¦
- 2012

We use optimism to introduce generic asymptotically optimal reinforcement learning agents. They achieve, with an arbitrary finite or compact class of environments, asymptotically optimal behavior.â€¦ (More)

- Gabriel Dulac-Arnold, Richard Evans, Peter Sunehag, Ben Coppin
- ArXiv
- 2015

Being able to reason in an environment with a large number of discrete actions is essential to bringing reinforcement learning to a larger class of problems. Recommender systems, industrial plantsâ€¦ (More)

- Peter Sunehag, Marcus Hutter
- AGI
- 2012

We consider extending the AIXI agent by using multiple (or even a compact class of) priors. This has the benefit of weakening the conditions on the true environment that we need to prove asymptoticâ€¦ (More)

- Peter Sunehag, Marcus Hutter
- Algorithmic Probability and Friends
- 2011

We identify principles characterizing Solomonoff Induction by demands on an agentâ€™s external behaviour. Key concepts are rationality, computability, indifference and time consistency. Furthermore, weâ€¦ (More)