Guiding High-Performance SAT Solvers with Unsat-Core Predictions

Selsam, Daniel; Bjørner, Nikolaj

doi:10.1007/978-3-030-24258-9_24

Cited by 62 publications

(68 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the opposite end, the valuable insights and knowledge gained from the use of ML in EDA flow can also be leveraged to aid adversarial attacks. For example, to improve the search engine for solving EDA problems, NeuroSAT [221] trains a GNN to classify Boolean Satisfiability (SAT) problems and it is simplified in [222] to guide the search process of an existing SAT solver. However, SAT solver can also be utilized as an oracle-guided attack to break logic locking by finding discriminative input patterns to quickly prune the search space of the secret key.…”

Section: On Trust Of ML For Electronic Design Automationmentioning

confidence: 99%

Two Sides of the Same Coin: Boons and Banes of Machine Learning in Hardware Security

Liu

Chang

Wang

et al. 2021

IEEE J. Emerg. Sel. Topics Circuits Syst.

View full text Add to dashboard Cite

The last decade has witnessed remarkable research advances at the intersection of machine learning (ML) and hardware security. The confluence of the two technologies has created many interesting and unique opportunities, but also left some issues in their wake. ML schemes have been extensively used to enhance the security and trust of embedded systems like hardware Trojans and malware detection. On the other hand, ML-based approaches have also been adopted by adversaries to assist side-channel attacks, reverse engineer integrated circuits and break hardware security primitives like Physically Unclonable Functions (PUFs). Deep learning is a subfield of ML. It can continuously learn from a large amount of labeled data with a layered structure. Despite the impressive outcomes demonstrated by deep learning in many application scenarios, the dark side of it has not been fully exposed yet. The inability to fully understand and explain what has been done within the super-intelligence can turn an inherently benevolent system into malevolent. Recent research has revealed that the outputs of Deep Neural Networks (DNNs) can be easily corrupted by imperceptibly small input perturbations. As computations are brought nearer to the source of data creation, the attack surface of DNN has also been extended from the input data to the edge devices. Accordingly, due to the opportunities of MLassisted security and the vulnerabilities of ML implementation, in this paper, we will survey the applications, vulnerabilities and fortification of ML from the perspective of hardware security. We will discuss the possible future research directions, and thereby, sharing a roadmap for the hardware security community in general.

show abstract

Section: On Trust Of ML For Electronic Design Automationmentioning

confidence: 99%

Two Sides of the Same Coin: Boons and Banes of Machine Learning in Hardware Security

Liu

Chang

Wang

et al. 2021

IEEE J. Emerg. Sel. Topics Circuits Syst.

View full text Add to dashboard Cite

show abstract

“…Both Li, Chen, and Koltun (2018) and Selsam and Bjørner (2019) used graphical neural networks to learn heuristics for guiding SAT solvers.…”

Section: Related Workmentioning

confidence: 99%

Predicting Propositional Satisfiability via End-to-End Learning

Cameron

Chen

Hartford

et al. 2020

AAAI

View full text Add to dashboard Cite

Strangely enough, it is possible to use machine learning models to predict the satisfiability status of hard SAT problems with accuracy considerably higher than random guessing. Existing methods have relied on extensive, manual feature engineering and computationally complex features (e.g., based on linear programming relaxations). We show for the first time that even better performance can be achieved by end-to-end learning methods — i.e., models that map directly from raw problem inputs to predictions and take only linear time to evaluate. Our work leverages deep network models which capture a key invariance exhibited by SAT problems: satisfiability status is unaffected by reordering variables and clauses. We showed that end-to-end learning with deep networks can outperform previous work on random 3-SAT problems at the solubility phase transition, where: (1) exactly 50% of problems are satisfiable; and (2) empirical runtimes of known solution methods scale exponentially with problem size (e.g., we achieved 84% prediction accuracy on 600-variable problems, which take hours to solve with state-of-the-art methods). We also showed that deep networks can generalize across problem sizes (e.g., a network trained only on 100-variable problems, which typically take about 10 ms to solve, achieved 81% accuracy on 600-variable problems).

show abstract

“…6, the network is not only fine-tuned using single-bit supervision, but it also followed by training with a single set of hyper-parameters as a coarse heuristic that broadly assigns higher CR to selected variables. This coarse learning also helps us to have more reliable (with higher confidence) predictions for UNSAT problems [10].…”

Section: Training Dataset Generationmentioning

confidence: 99%