Accepted paper at SAC '26

Shurok’s latest thesis contribution has been accepted at SAC, a primary gathering forum for applied computer scientists, computer engineers, software engineers, and application developers from around the world, in the Machine Learning and Its Application (MLA) track, which will take place this year in Thessaloniki (Greece), from March 23rd to 27th.

Shurok’s work, entitled SMART: Scalable Mitigation Architecture using Reinforcement learning for DDoS Attack Traffic in SDN Environments, will be presented on Wednesday 25th March during the MLA-5 session between 11:00 and 12:30 in Grace Hall C.

The paper improves on Shurok’s previous proposals dealing with adaptive DDoS mitigation leveraging a Reinforcement Learning agent. In particular, SMART attempts to counter a wide range of flooding-based attacks across diverse protocols while preserving QoS for legitimate users. Our approach introduces a group-based state projection mechanism to reduce input dimensionality, thereby minimizing computational overhead without sacrificing essential network insights. To ensure adaptability across diverse network topologies, we employ modular neural architectures with permutation-invariant and permutation-equivariant functions, enabling the model to generalize across varying network sizes and entry point configurations without retraining. Experimental evaluations in an emulated SDN environment demonstrate the framework’s robustness, scalability, and efficiency in maintaining service availability for legitimate users.