Joel Veness
E774594
Joel Veness is a computer scientist and researcher known for his work in artificial intelligence and algorithmic information theory, including collaborations with Marcus Hutter.
All labels observed (1)
| Label | Occurrences |
|---|---|
| Joel Veness canonical | 1 |
Statements (42)
| Predicate | Object |
|---|---|
| instanceOf |
computer scientist
ⓘ
researcher ⓘ |
| affiliation | DeepMind NERFINISHED ⓘ |
| coauthoredWith |
David Silver
NERFINISHED
ⓘ
Demis Hassabis NERFINISHED ⓘ Karen Simonyan NERFINISHED ⓘ Kee Siong Ng NERFINISHED ⓘ Koray Kavukcuoglu NERFINISHED ⓘ Marcus Hutter NERFINISHED ⓘ Michael Bowling NERFINISHED ⓘ Simon Schmitt NERFINISHED ⓘ Tom Schaul NERFINISHED ⓘ Wojciech Czarnecki NERFINISHED ⓘ |
| collaboratedWith | Marcus Hutter NERFINISHED ⓘ |
| countryOfCitizenship | Australia ⓘ |
| educatedAt | University of Alberta NERFINISHED ⓘ |
| employer | DeepMind NERFINISHED ⓘ |
| fieldOfWork |
algorithmic information theory
ⓘ
artificial intelligence ⓘ machine learning ⓘ reinforcement learning ⓘ |
| hasGivenTalkAt |
AI and RL seminars
ⓘ
NIPS workshops NERFINISHED ⓘ |
| knownFor |
applications of algorithmic information theory to AI
ⓘ
research on universal prediction and planning ⓘ work on Monte Carlo search methods ⓘ work on reinforcement learning agents ⓘ |
| language | English ⓘ |
| notableConcept | practical approximations to AIXI ⓘ |
| notableWork |
A Monte Carlo AIXI Approximation
NERFINISHED
ⓘ
Reinforcement Learning via AIXI Approximation NERFINISHED ⓘ |
| notableWorkArea | universal artificial intelligence ⓘ |
| publishedIn |
Advances in Neural Information Processing Systems
NERFINISHED
ⓘ
International Conference on Machine Learning NERFINISHED ⓘ Journal of Artificial Intelligence Research NERFINISHED ⓘ Proceedings of the AAAI Conference on Artificial Intelligence NERFINISHED ⓘ |
| researchInterest |
general reinforcement learning
ⓘ
planning under uncertainty ⓘ universal sequence prediction ⓘ |
| supervisedBy | Michael Bowling NERFINISHED ⓘ |
| worksOn |
game-playing AI agents
ⓘ
scalable reinforcement learning algorithms ⓘ |
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.