adversarial-learning graph-neural-networks logic-locking logic-synthesis machine-learning-attack simulated-annealing electronics-design-automation graph-machine-learning adversarial-attacks

ALMOST: Adversarial Learning to Mitigate Oracle-less ML Attacks via Synthesis Tuning

Contact Info

This repository contains implementation of the experiments in the following paper:

A. B. Chowdhury, L. Alrahis, L. Collini, J. Knechtel, R. Karri, S. Garg, O. Sinanoglu and B. Tan , "ALMOST: Adversarial Learning to Mitigate Oracle-less ML Attacks via Synthesis Tuning" to appear in DAC 2023

Pre-print

The repository is raw implementation during DAC submission. It needs refactoring. Stay tuned for more updates.

About

ALMOST: Adversarial Learning to Mitigate Oracle-less ML Logic Locking Attacks via Synthesis Tuning

adversarial-learning graph-neural-networks logic-locking logic-synthesis machine-learning-attack simulated-annealing electronics-design-automation graph-machine-learning adversarial-attacks

Languages

Language:Verilog 99.9%Language:Python 0.1%Language:Perl 0.0%Language:Scala 0.0%Language:Shell 0.0%