CSE 522
The following is an incomplete and somewhat random collection of resources and papers
that could be used for the project.
Feel free to follow your own passion as long as
it "fits" the course.
General paper sources
Theoretical papers from the following conferences: COLT, ICML, NIPS, SODA, FOCS, STOC, EC, ITCS
Papers by some leading researchers on related topics (this is a very incomplete list):
Papers referenced in related courses; e.g.,
Peter Bartlett's course
Generalization bounds and sample complexity
Online learning
- The computational power of optimization in online learning, Hazan and Koren.
- Kernel-based methods for bandit convex optimization, Bubeck, Eldan, Lee
- Regularization techniques for learning with matrices, Kakade, Shalev-Shwartz, Tewari.
- The best of both worlds: stochastic and adversarial bandits, Bubeck and Slivkins.
- Know your customer - multi-armed bandits with capacity constraints, Johari, Kamble, Kanoria.
- The importance of exploration in online marketplaces, Banerjee, Zhou and Johari.
- Approximation algorithms for restless bandits, Guha, Munagala, Shi.
- Taming the monster: A fast and simple algorithm for contextual bandits
- Analysis of Thompson Sampling for multi-armed bandits, Agrawal and Goyal.
- Projection-free online learning, Hazan, Kale
- Oracle-efficient online learning and auctions design by Dudik et al
Optimization related
Learning with malicious, untrusted, costly or manipulated data, game theoretic aspects
- Algorithmic game theory and data science references, by Vasilis Syrgkanis
- Oracle efficient online learning and auction design Dudik et al.
- Fast convergence of regularized learning in games, Syrgkanis et al.
- Incentive compatible regression learning Dekel et al
- Truthful univariate estimators, Caragiannis et al
- Truthful linear regression, Cummings et al
- Accuracy for sale, Cummings et al
- Double or nothing: multiplicative incentive mechanisms for crowdsourcing Shah et al
- Efficient PAC learning from the crowd Awasthi et al.
- No oops, you won't do it again: mechanisms for self-correction in crowdsourcing, by Shah et al
- Learning from untrusted data, Charikar, Steinhardt and Valiant.
- Low-cost learning via active data procurement Abernethy, Chen, Ho and Waggoner.
- Optimum statistical estimation with strategic data resources Cai, Daskalakis, Papadimitriou.
- A bandit framework for strategic regression, Liu and Chen.
- Provably manipulation-resistant reputation systems, Christiano.
- Bayesian exploration: incentivizing exploration in Bayesian games , Mansour et al
- Incentivizing exploration Frazier et al.
- Watch and learn: optimizing from revealed preferences feedback, Roth et al
- Peer prediction with heterogeneous users Agarwal et al
- Avoiding impostors and delinquents: adversarial crowdsourcing and peer prediction, by Steinhardt et al
- Regularized minimax conditional entropy for crowdsourcing
Other