Skip to content
STIMSMITH

Reinforcement learning (RL)

Technique
First seen 6/24/2026
Last seen 7/14/2026
Evidence 6 chunks

NEIGHBORHOOD

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

explore full graph →

RELATIONSHIPS

7 connections
ChatFuzz ← uses 96% 1e
RL optimizes the input generator based on coverage rewards.
Beyond Random Inputs: A Novel ML-Based Hardware Fuzzing ← uses 95% 1e
RL optimizes input generation through coverage-based rewards.
Reward Shaping uses → 100% 1e
The RL agent in the proposed framework relies on reward shaping to guide coverage-directed exploration.
Coverage-Directed Generation (CDG) implements → 100% 1e
The RL agent maximizes coverage discovery, implementing CDG in the proposed framework.
Coverage Closure uses → 100% 1e
The RL agent drives coverage closure by refining its policy based on coverage feedback.
Hierarchical Learning ← extends 85% 1e
Hierarchical learning extends RL by training customized sub-policies for different design modules.
Explainable AI (XAI) ← uses 85% 1e
XAI techniques would be used to explain the RL agent's constraint selection decisions.