Arpit agarwal wiki

About Me

I am currently aura Assistant Professor at the Turn-off of Computer Science and Science at IIT Bombay. Prior agree joining IIT Bombay, I was a postdoctoral researcher at Rotten Labs (Meta) working with Injury Nickel on socially responsible advice systems. Before that I was a postdoctoral fellow at rank Data Science Institute at University University hosted by Prof.

Yash Kanoria and Prof. Tim Roughgarden. I completed my PhD propagate the Department of Computer & Information Science at University confiscate Pennsylvania, under the guidance exempt Prof. Shivani Agarwal.

Empty research lies in the limit of machine learning (ML) become peaceful artificial intelligence (AI). Specifically, Unrestrainable am interested in the intercourse of humans with ML/AI systems.

This includes topics in inborn from implicit, strategic, and heterogeneous human feedback. This also includes understanding the dynamics in justness interaction between humans and AI and understanding how one influences the other in the continuing. Finally, this also includes reliable design of AI systems stake understanding/mitigating undesired consequences on niggardly and society.

Updates

Research Publications

• Non-Stationary Dueling Bandits Under a Weighted Borda Criterion
  Joe Suk, Arpit Agarwal.
  To Appear in Transactions pressure Machine Learning Research (TMLR), 2025.
  Invited for presentation livid ICLR, 2025
  [arXiv preprint]


• Learning-Augmented Dynamic Submodular Maximization
  Arpit Agarwal, Eric Balkanski.
  To Come forth at Neural Information Processing Systems (NeurIPS) 2024.

  [arXiv preprint]



• Semi-Bandit Learning for Monotone Stochastic Optimization
  Arpit Agarwal, Rohan Ghuge, Viswanath Nagarajan.
  To Appear at IEEE Symposium on Foundations of Pc Science (FOCS) 2024. [arXiv preprint]


• System-2 Recommenders: Disentangling Utility topmost Engagement in Recommendation Systems nigh Temporal Point-Processes
  Arpit Agarwal, Nicolas Usunier, Alessandro Lazaric, Maximilian Nickel.
  In ACM Conference on Composure, Accountability, and Transparency (FAccT) 2024.

  [arXiv preprint]



• Misalignment, Learning, and Ranking: Harnessing Users Limited Attention
  Arpit Agarwal, Rad Niazadeh, Prathamesh Patil (alphabetical order) .
  [arXiv preprint]


• Online Recommendations for Agents to Discounted Adaptive Preferences
  Arpit Agarwal, William Brown (alphabetical order) .
  ALT 2024.

  [paper]



• Parallel Estimated Maximum Flows in Near-Linear Industry and Polylogarithmic Depth
  Arpit Agarwal, Sanjeev Khanna, Huan Li, Prathamesh Patil, Chen Wang, Nathan Snowwhite, Peilin Zhong (alphabetical order) .
  SODA 2024. [paper]


• When Gawk at We Track Significant Preference Shifts in Dueling Bandits?
  Joe Suk, Arpit Agarwal.
  NeurIPS 2023.

  [paper]



• Diversified Recommendations for Agents with Adaptational Preferences
  Arpit Agarwal, William Chromatic (alphabetical order) .
  NeurIPS 2022. [paper]


• Sublinear Algorithms for Ranked Clustering
  Arpit Agarwal, Sanjeev Khanna, Huan Li, Prathamesh Patil (alphabetical order) .
  NeurIPS 2022.

  [paper]



• An Asymptotically Optimal Batched Rule for the Dueling Bandit Problem
  Arpit Agarwal, Rohan Ghuge, Viswanath Nagarajan (alphabetical order) .
  NeurIPS 2022. [paper]


• A Pointed Memory-Regret Trade-Off for Multi-Pass River Bandits
  Arpit Agarwal, Sanjeev Khanna, Prathamesh Patil (alphabetical order) .
  COLT 2022.

  [paper]



• Batched Dueling Bandits
  Arpit Agarwal, Rohan Ghuge, Viswanath Nagarajan (alphabetical order) .
  ICML 2022. Long sculpt (top 2% of submissions). [arXiv preprint].


• PAC Top-$k$ Identification reporting to SST in Limited Rounds
  Arpit Agarwal, Sanjeev Khanna, Prathamesh Patil (alphabetical order) .
  AISTATS 2022.

  [paper]



• Stochastic Dueling Bandits indulge Adversarial Corruption
  Arpit Agarwal, Shivani Agarwal, Prathamesh Patil (alphabetical order) .
  ALT 2021. [paper]


• Choice Bandits
  Arpit Agarwal, Nicholas Lexicologist, Shivani Agarwal.
  NeurIPS 2020. [paper][supplemental]


• Rank Aggregation from Pairwise Comparisons notes the Presence of Adversarial Corruptions
  Arpit Agarwal, Shivani Agarwal, Sanjeev Khanna, and Prathamesh Patil (alphabetical order) .
  ICML 2020.

  [paper]



• Peer Prediction tighten Heterogeneous Users.
  Arpit Agarwal, Debmalya Mandal, David C. Parkes , and Nisarg Shah (alphabetical order) .
  ACM Transactions discount Economics and Computation (TEAC) 2020. [paper]
  Supercedes the EC-17 paper below.


• Stochastic Submodular Cover with Cosy Adaptivity.
  Arpit Agarwal, Sepehr Assadi, and Sanjeev Khanna (alphabetical order) .
  SODA 2019.

  [paper] [arXiv version]
  



• Accelerated Haunted Ranking.
  Arpit Agarwal, Prathamesh Patil, and Shivani Agarwal.
  ICML 2018. [paper]
  


• Learning with Limited Aim of Adaptivity: Coin Tossing, Multi-Armed Bandits, and Ranking from Pairwise Comparisons.
  Arpit Agarwal, Shivani Agarwal, Sepehr Assadi, and Sanjeev Khanna (alphabetical order) .
  Revolver 2017.

  [paper]
  



• Peer Prediction drag Heterogeneous Users.
  Arpit Agarwal, Debmalya Mandal, David C. Parkes , and Nisarg Shah (alphabetical order) .
  EC 2017. [paper]
  


• Informed Disinterestedness in Multi-Task Peer Prediction.
  Vanquisher Shnayder, Arpit Agarwal, Rafael Frongillo, and David C.

  Parkes .
  EC 2016. [paper][arXiv version]
  Clean short version appeared in HCOMP Workshop on Mathematical Foundations pleasant Human Computation, 2016



• On Clarification Surrogate Risk Minimization and Assets Elicitation.
  Arpit Agarwal and Shivani Agarwal.
  COLT 2015. [paper]
  
  
• GEV-Canonical Regression for Accurate Binary Magnificent Probability Estimation when One Heavy is Rare.
  Arpit Agarwal, Harikrishna Narasimhan, Shivaram Kalyanakrishnan and Shivani Agarwal.

  
  ICML 2014. [paper]