Anuran Makur

Anuran Makur is an active Assistant Professor at Purdue University with dual appointments in the Department of Computer Science (College of Science) and the Elmore Family School of Electrical and Computer Engineering (College of Engineering). He is affiliated with the Institute for Control, Optimization and Networks (ICON) and teaches foundational courses in machine learning and data science.

His educational background includes a B.S. in Electrical Engineering and Computer Sciences from UC Berkeley (2013, summa cum laude), an S.M. in Electrical Engineering and Computer Science from MIT (2015), and a Sc.D. from MIT (2019).

• B.S., UC Berkeley, 2013
• S.M., MIT, 2015
• Sc.D., MIT, 2019

Makur's research bridges theoretical machine learning, information theory, and applied probability. Key interests include ranking/preference learning, optimization for ML, non-parametric inference, information measures, permutation channel limits, broadcasting on graphs, and reliable computation. His work emphasizes fundamental theoretical limits and mathematical rigor in complex systems.
Recent publications reveal strong trends in statistical learning theory (40%), information-theoretic methods (35%), and networked systems (25%), with growing emphasis on privacy-aware inference and high-dimensional statistics.

His scientific achievements are recognized by prestigious awards:

• Arthur M. Hopkin Award (UC Berkeley, 2013)
• Ernst A. Guillemin Master's Thesis Award (MIT, 2015)
• Jin Au Kong Doctoral Thesis Award (MIT, 2020)
• Thomas M. Cover Dissertation Award (IEEE, 2021)
• NSF CAREER Award (2023)

While specific advising details aren't public, his research leadership is evident through ICON affiliation and collaborations with MIT's LIDS/IDSS groups. The NSF CAREER grant supports his work on information-theoretic foundations of machine learning. He maintains active roles in theoretical computer science and information theory communities through conference organization and editorial work.

Makur leads research within ICON, focusing on control-theoretic approaches to networked learning systems. His work integrates probabilistic modeling with optimization theory, particularly for distributed inference and networked decision-making under uncertainty.