Dmitrii Krasheninnikov

Artificial Intelligence · 2022–2023

A researcher in the Science corpus: 4 papers; trusts 2 researchers and is trusted by 0 researchers here, as inferred from the citation graph.

Subfields

• Artificial Intelligence — 167 citations into it
• Reinforcement Learning — 74 citations into it

Whom Dmitrii Krasheninnikov trusts overall

Their strongest outbound trust, strongest first, inferred from their citations.

1. Brian D. Ziebart
2. Andrew Y. Ng

Papers from Dmitrii Krasheninnikov's perspective

Ordered by how strongly the researchers Dmitrii Krasheninnikov trusts endorse each paper. Same corpus, a different order for every researcher.

1. Playing Atari with Deep Reinforcement Learning (arXiv.org)
2. Algorithms for Inverse Reinforcement Learning
3. Maximum Entropy Inverse Reinforcement Learning
4. Deep Reinforcement Learning from Human Preferences (Neural Information Processing Systems)
5. Mastering the game of Go with deep neural networks and tree search
6. Concrete Problems in AI Safety
7. Training language models to follow instructions with human feedback (Neural Information Processing Systems)
8. AI safety via debate
9. Learning to summarize from human feedback
10. Scaling Laws for Neural Language Models (arXiv.org)

Their papers

• Defining and Characterizing Reward Hacking (2022) — 169 citations
• Harms from Increasingly Agentic Algorithmic Systems (2023) — 259 citations
• Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback (2023) — 934 citations
• Defining and Characterizing Reward Gaming (2022) — 379 citations