Dr Robert Loftin

Department of Computer Science

Lecturer in Machine Learning

R.Loftin@sheffield.ac.uk
+44 114 215 7542

Regent Court (DCS)

Full contact details

Dr Robert Loftin
Department of Computer Science
Regent Court (DCS)
211 Portobello
Sheffield
S1 4DP

Profile: Robert received his PhD in Computer Science from North Carolina State University in 2019, under the supervision of Dave Roberts. His dissertation examined the types of latent knowledge conveyed by human-provided feedback and demonstrations, and developed interactive learning algorithms which leverage models of human behavior to extract this information. He received his Bachelor's in Computer Science from Georgia Tech in 2011. After completing his PhD, he did a two-year post-doc with Microsoft Research Cambridge, exploring the use of Reinforcement Learning and Interactive Learning in commercial game development. He also completed a post-doc at TU Delft with Frans Oliehoek, where he focused on applying game theory and multi-agent reinforcement learning to human-AI cooperation. Robert is currently a Lecturer in Machine Learning at the University of Sheffield.

Research interests

- Ad Hoc Human-AI Cooperation

- Human-AI Alignment

- Human-in-the-loop Machine Learning

- Deep Reinforcement Learning

- Multi-Agent RL

- Human-Robot Interaction

- AI for Games

Publications

Peng B, MacGlashan J, Loftin R, Littman ML, Roberts DL & Taylor ME (2018) Curriculum Design for Machine Learners in Sequential Decision Tasks. IEEE Transactions on Emerging Topics in Computational Intelligence, 2(4), 268-277.
Loftin R, Peng B, MacGlashan J, Littman ML, Taylor ME, Huang J & Roberts DL (2016) Learning behaviors via human-delivered discrete feedback: modeling implicit feedback strategies to speed up learning. Autonomous Agents and Multi-Agent Systems, 30(1), 30-59.

Aydeniz AA, Loftin R & Tumer K (2023) Novelty Seeking Multiagent Evolutionary Reinforcement Learning. Proceedings of the Genetic and Evolutionary Computation Conference
Loftin R & Oliehoek FA (2022) On the Impossibility of Learning to Cooperate with Adaptive Partner Strategies in Repeated Games. Proceedings of Machine Learning Research, Vol. 162 (pp 14197-14209)
Loftin R, Saha A, Devlin S & Hofmann K (2021) Strategically Efficient Exploration in Competitive Multi-agent Reinforcement Learning. Proceedings of Machine Learning Research, Vol. 161 (pp 1587-1596)
Loftin R, Saha A, Devlin S & Hofmann K (2021) Strategically Efficient Exploration in Competitive Multi-agent Reinforcement Learning. 37th Conference on Uncertainty in Artificial Intelligence, UAI 2021 (pp 1587-1596)
Ciosek K, Loftin R, Vuong Q & Hofmann K (2019) Better exploration with optimistic actor-critic. Advances in Neural Information Processing Systems, Vol. 32
Smith E, Loftin R, Murphy-Hill E, Bird C & Zimmermann T (2013) Improving developer participation rates in surveys. 2013 6th International Workshop on Cooperative and Human Aspects of Software Engineering (CHASE), 25 May 2013 - 25 May 2013.
Loftin R, Çelikok MM, van Hoof H, Kaski S & Oliehoek FA () Uncoupled Learning of Differential Stackelberg Equilibria with Commitments. Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2024). Auckland, New Zealand, 6 May 2024 - 6 May 2024.

Loftin R, Çelikok MM, Hoof HV, Kaski S & Oliehoek FA (2023) Uncoupled Learning of Differential Stackelberg Equilibria with Commitments.