Network Algorithms and Distributed Optimization

We design and analyze approximation algorithms for NP-hard network optimization problems. These algorithms demonstrate the power of randomness for overcoming provably difficult tasks. Our theoretical results are supported by simulations to showcase their practicality. Recently, we have been investigating the application of reinforcement learning techniques to network optimization problems that include local restrictions.

Refereed Conference Proceedings

• Distributed Online Rollout for Multivehicle Routing in Unmapped Environments. Jamison W. Weber, Dhanush R. Giriyan, Devendra R. Parkar, Dimitri P. Bertsekas, Andréa W. Richa. AAMAS ’24: Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems, May 2024, Pages 1910 – 1918.
• Improved Throughput for All-or-Nothing Multicommodity Flows with Arbitrary Demands. Anya Chaturvedi, Chandra Chekuri, Andréa W. Richa, Stefan Schmid, Matthias Rost, Jamison Weber. 39th International Symposium on Computer Performance, Modeling, Measurements and Evaluation, 2021.
• A Constant Approximation for Maximum Throughput Multicommodity Routing And Its Application to Delay-Tolerant Network Scheduling. Mengxue Liu, Andréa W. Richa, Matthias Rost, Stefan Schmid. IEEE Conference on Computer Communications (INFOCOM 2019), pp. 46-54, 2019.

Conference Talks

• Improved Throughput for All-or-Nothing Multicommodity Flows with Arbitrary Demands. Anya Chaturvedi, Jamison Weber. 39th International Symposium on Computer Performance, Modeling, Measurements and Evaluation, 2021.