Contact Information:
Coordinated Science Laboratory
University of Illinois at Urbana-Champaign
1308 W Main Street, CSL-128
Urbana, IL 61801

Email: scayci[at]illinois[dot]edu

I am an NSF TRIPODS Postdoctoral Fellow at the University of Illinois at Urbana-Champaign. I am working with Prof. R. Srikant (UIUC) and Prof. Niao He (ETH Zurich). My research interests broadly lie in reinforcement learning, online learning and deep learning theory.

The ultimate goal of my research is to develop efficient, robust and scalable data-driven decision-making methods that achieve provably good performance for large-scale systems. To that end, I take a principled approach to the design and analysis of reinforcement learning algorithms. The application areas of my research include stochastic networks, cloud computing, algorithmic fairness and autonomous systems.

My curriculum vitae can be found here.



Selected Recent Papers

A Lyapunov-Based Methodology for Constrained Optimization with Bandit Feedback
Semih Cayci, Yilin Zheng and Atilla Eryilmaz
To appear in the Proc. of AAAI 2022, arXiv:2106.05165
[paper]

Sample Complexity and Overparameterization Bounds for Temporal Difference Learning with Neural Network Approximation
Semih Cayci, Siddhartha Satpathi, Niao He, R. Srikant
Short version appeared at ICML Workshop on Overparameterization, 2021
[paper]

Group-Fair Online Allocation in Continuous Time
Semih Cayci, Swati Gupta, Atilla Eryilmaz
Advances in Neural Information Processing Systems 33, NeurIPS 2020
[paper]

Budget-Constrained Bandits over General Cost and Reward Distributions
Semih Cayci, Atilla Eryilmaz, R. Srikant
International Conference on Artificial Intelligence and Statistics, AISTATS 2020.
[paper][supp][slides]

Learning to Control Renewal Processes with Bandit Feedback
Semih Cayci, Atilla Eryilmaz, R. Srikant
ACM SIGMETRICS 2019
[paper]