Contact:

About:

I am ARC Postdoctoral Fellow at Georgia Tech hosted by Jake Abernethy, Vidya Muthukumar, and Juba Ziani.

My research interests are at the intersection of online learning, optimization, and algorithmic game theory. My current and recent work has focused on developing principled theoretical foundations for solving online learning tasks, particularly in multi-agent and strategic settings.

Previously, I was a postdoctoral researcher at SUTD in Singapore working with Georgios Piliouras and Antonios Varvitsiotis. I received my PhD in computer science at Yale in May 2024, where I was advised by James Aspnes.

Recent News:

• August 2026: started as an ARC postdoc at Georgia Tech.
• September 2025: along with Anas Barakat and Joseph Sakos, designed and taught a new graduate-level course on online learning and learning in games at SUTD.
• Summer 2023 and 2024: Visited IST Austria, hosted by Dan Alistarh and Krish Chatterjee.

Preprints:

• When and Why is Optimistic Multiplicative Weights Slow? The Geometry of Energy Dissipation.
  John Lazarsfeld, Anas Barakat, Georgios Piliouras, Antonios Varvitsiotis, and Andre Wibisono
  May 2026 -- preprint
  [ArXiv]
• Online Multi-Agent Control with Adversarial Disturbances
  Anas Barakat, John Lazarsfeld, Georgios Piliouras, and Antonios Varvitsiotis
  May 2025 -- preprint
  [ArXiv]

Publications:

• Optimistic Online Learning in Symmetric Cone Games
  Anas Barakat, Wayne Lin, John Lazarsfeld, and Antonios Varvitsiotis
  TMLR 2026
  [ArXiv]
• Optimism Without Regularization: Constant Regret in Zero-Sum Games
  John Lazarsfeld, Georgios Piliouras, Ryann Sim, and Stratis Skoulakis
  NeurIPS 2025
  [ArXiv]
• Fast and Furious Symmetric Learning in Zero-Sum Games: Gradient Descent as Fictitious Play
  John Lazarsfeld, Georgios Piliouras, Ryann Sim, and Andre Wibisono
  COLT 2025
  [ArXiv] [COLT slides]
• Game Dynamics and Equilibrium Computation in the Population Protocol Model
  Dan Alistarh, Krishnendu Chatterjee, Mehrdad Karrabi, and John Lazarsfeld
  PODC 2024
  [ArXiv]
• Simple Opinion Dynamics for No-Regret Learning
  John Lazarsfeld and Dan Alistarh
  OPT workshop @ NeurIPS 2023
  [ArXiv]
• Fast Convergence of k-Opinion Undecided State Dynamics in the Population Protocol Model
  Talley Amir, James Aspnes, Petra Berenbrink, Felix Biermeier, Christopher Hahn, Dominik Kaaser, and John Lazarsfeld
  PODC 2023
  [ArXiv]
• Differentially Private Maximal Information Coefficients
  John Lazarsfeld, Aaron Johnson, and Emmanuel Adeniran
  ICML 2022
  [ArXiv]
• Consistency of the Maximal Information Coefficient Estimator
  John Lazarsfeld and Aaron Johnson
  Technical Report, July 2021
  Corrects an error in previous work of Reshef et al. (JMLR, 2016)
  [ArXiv]
• Approximate Majority With Catalytic Inputs
  OPODIS 2020
  Talley Amir, James Aspnes, and John Lazarsfeld
  [ArXiv]
• Majority Vote Cascading: A Semi-Supervised Framework for Improving Protein Function Prediction
  John Lazarsfeld, Jonathan Rodríguez, Mert Erden, Yuelin Liu, and Lenore Cowen
  ACM BCB 2019 - Best Poster Award
  [Conf. Proceedings] [Code and Data]

Teaching:

• At SUTD: Course Designer and Co-Instructor
  SUTD 40.616: Special Topics in Games, Learning, and Optimization (Fall 2025)
  Course website and notes: learning-in-games.github.io
  


• At Yale: Graduate Teaching Fellow
  CPSC 569: Randomized Algorithms (2024)
  CPSC 365: Introduction to Algorithms (2021, 2022)
  CPSC 202: Mathematical Tools for Computer Science (2020, 2021)

Other/Service: