Publications

Adarsh Barik, Anand Krishna, Vincent Y. F. Tan

UAI 2026

A non-convex but invex loss (LEARN) that tolerates an unknown number of corrupted rounds, with tight dynamic regret guarantees and no Lipschitz assumption.

Adarsh Barik, Anand Krishna, Vincent Y. F. Tan

IEEE Transactions on Information Theory

Recovers a signal from magnitude-only measurements under arbitrary corruption, with nearly linear sample complexity and no spectral initialization.

Anand Krishna, Philips George John, Adarsh Barik, Vincent Y. F. Tan

AAAI 2025

A single UCB-based algorithm that achieves $p$-mean regret bounds across the whole fairness–efficiency spectrum, recovering average and Nash regret as special cases.

Siddharth Barman, Anand Krishna, Pooja Kulkarni, Shivika Narang

ITCS 2024

Breaks the $O(n)$-approximation barrier for Nash social welfare under XOS valuations, using demand and XOS oracles.

Siddharth Barman, Anand Krishna, Y. Narahari, Soumyarup Sadhukhan

WINE 2022

A polynomial-time $(18+o(1))$-approximation for Nash social welfare in coverage problems, with a matching APX-hardness result.

Siddharth Barman, Anand Krishna, Y. Narahari, Soumyarup Sadhukhan

IJCAI 2022

Under dichotomous valuations, envy-freeness is reachable with a per-agent subsidy of only $0$ or $1$ — a linear-factor improvement on general monotone valuations.

Siddharth Barman, Umang Bhaskar, Anand Krishna, Ranjani G. Sundaram

ESA 2020

An $8n$-approximation for $p$-mean welfare under subadditive valuations, with a matching lower bound showing the guarantee is essentially tight.