Published in International Journal of Information Technology, 2026
This paper analyses neural networks deployed to play games to uncover bugs in the games.
Recommended citation: Rani, G., Pandey, U., Wagde, A.A. et al. A deep reinforcement learning technique for bug detection in video games. Int. j. inf. tecnol. 15, 355–367 (2023). https://doi.org/10.1007/s41870-022-01047-z
Download Paper
Published in OPT 2025: Optimization for Machine Learning, NeurIPS 2025 Workshop, 2026
This paper describes a technique to reduce combinatorial-bandit problems to bandits with pairwise preferences and solve them efficiently.
Recommended citation: Wagde, A., Saha, A., Efficient Algorithms for Combinatorial-Bandits with Monotonicity. NeurIPS OPTML (2025). https://openreview.net/forum?id=D3drIoEW5B
Download Paper
Published in AISTATS 26, 2026
This paper formulates LLM-as-a-judge evaluation as a variance-adaptive resource allocation problem, using multi-armed bandits to minimize estimation error under a fixed budget.
Recommended citation: Saha, A., Wagde, A., Kveton, B., LLM-as-Judge on a Budget. arXiv preprint arXiv:2602.15481 (2026).
Download Paper