Dharshan Kumaran
Google DeepMind (United Kingdom)
2 papers in the Anchorcite directory, published between 2017 and 2018, cited 14,853 times in total.
ORCID profileResearch topics
- Reinforcement Learning in Robotics · 2 papers
- Neural Networks and Applications · 1 paper
- Domain Adaptation and Few-Shot Learning · 1 paper
- Artificial Intelligence Applications · 1 paper
- Artificial Intelligence in Games · 1 paper
Frequent co-authors
- Demis Hassabis · 2 papers
- James Kirkpatrick · 1 paper
- Razvan Pascanu · 1 paper
- Neil C. Rabinowitz · 1 paper
- Joel Veness · 1 paper
- Andrei A. Rusu · 1 paper
- Kieran Milan · 1 paper
- John Quan · 1 paper
Papers
- Overcoming catastrophic forgetting in neural networksProceedings of the National Academy of Sciences · 2017 · 8,816 citations
The ability to learn tasks in a sequential fashion is crucial to the development of artificial intelligence. Until now neural networks have not been capable of this and it has been widely thought that catastrophic forgetting is an inevitable feature…
- A general reinforcement learning algorithm that masters chess, shogi, and Go through self-playScience · 2018 · 6,037 citations
The game of chess is the longest-studied domain in the history of artificial intelligence. The strongest programs are based on a combination of sophisticated search techniques, domain-specific adaptations, and handcrafted evaluation functions that have been refined by human experts over…
Metadata from OpenAlex (CC0).