Publications

Stochastic Domination of Gaussian Maxima: A Resolution to the Weak Simplex Conjecture

A. Mulgund

arXiv:2607.14087, July 2026

On loss landscapes of error-correcting codes: symmetries, linear codes and stationarity

A. Mulgund and R. Shekar

Submitted to Information Theory Workshop (ITW), May 2026

A Principled Analysis of Antidistillation Sampling

A. Mulgund and C. Pabbaraju

Under review, May 2026

Relating Misfit to Gain in Weak-to-Strong Generalization Beyond the Squared Loss

A. Mulgund and C. Pabbaraju

International Conference on Machine Learning (ICML), July 2025

Decomposing the Training of Deep Learned Turbo codes via a Feasible MAP Decoder

A. Mulgund, N. Devroye, G. Turán, and M. Žefran

International Symposium on Topics in Coding (ISTC), September 2023

Interpreting Training Aspects of Deep-Learned Error-Correcting Codes

N. Devroye, A. Mulgund, R. Shekhar, G. Turán, M. Žefran, and Y. Zhou

IEEE International Symposium on Information Theory (ISIT), June 2023

Evaluating interpretations of deep-learned error-correcting codes

A. Mulgund, R. Shekhar, N. Devroye, G. Turán, and M. Žefran

58th Annual Allerton Conference on Communication, Control, and Computing, September 2022

Interpreting Deep-Learned Error-Correcting Codes

N. Devroye, N. Mohammadi, A. Mulgund, H. Naik, R. Shekhar, G. Turán, Y. Wei, and M. Žefran

IEEE International Symposium on Information Theory (ISIT), June 2022