recurrent neural network-based constraint alteration

Technique

Recurrent neural network-based constraint alteration is a technique for simulation-based processor verification in which a recurrent neural network dynamically changes pseudorandom-generator constraints using coverage feedback from simulations of the design under verification.

First seen 5/26/2026

Last seen 5/26/2026

Evidence 2 chunks

Wiki v1

WIKI

Overview

Recurrent neural network-based constraint alteration is a technique proposed for simulation-based verification of processors. In the described verification flow, stimuli are generated by pseudorandom generators (PRGs), applied to processor inputs, and evaluated by monitoring achieved functional coverage to judge verification completeness.[C1]

The technique dynamically alters PRG constraints through a recurrent neural network (RNN). The RNN receives coverage feedback from simulation of the design under verification and uses that feedback in the constraint-alteration process.[C2]

READ FULL ARTICLE →

NEIGHBORHOOD

No graph connections found for this entity yet. It may appear in future ingestion runs.

explore full graph →

RELATIONSHIPS

1 connections

Automation of Processor Verification Using Recurrent Neural Networks ← introduces 100% 2e

The paper proposes a new technique that dynamically alters constraints for PRG via a recurrent neural network.

LINKED ENTITIES

1 links

Automation of Processor Verification Using Recurrent Neural Networks described_by The evidence identifies this paper title and its abstract describes the recurrent neural network technique for dynamically altering PRG constraints.

CITATIONS

7 sources

7 citations — click to expand

[1] Simulation-based processor verification commonly generates stimuli with pseudorandom generators, applies them to processor inputs, and monitors functional coverage to assess verification completeness. Automation of Processor Verification Using Recurrent Neural Networks

[2] The proposed technique dynamically alters PRG constraints via a recurrent neural network that receives coverage feedback from simulation of the design under verification. Automation of Processor Verification Using Recurrent Neural Networks

[3] Stimuli may be represented as bit vectors applied to processor input ports or as programs loaded directly into program memory. Automation of Processor Verification Using Recurrent Neural Networks

[4] The demonstration used processors provided by Codasip because their coverage state space is reasonably large and differs for various processor kinds. Automation of Processor Verification Using Recurrent Neural Networks

[5] The authors state that the presented techniques are widely applicable. Automation of Processor Verification Using Recurrent Neural Networks

[6] Experimental results showed faster coverage closure and isolation of a small set of high-coverage stimuli usable for regression tests. Automation of Processor Verification Using Recurrent Neural Networks

[7] The source paper is titled "Automation of Processor Verification Using Recurrent Neural Networks" and is listed as arXiv:1803.09810 / arXiv:1803.09810v1. Automation of Processor Verification Using Recurrent Neural Networks