Work

Explainable VQPG CartPole AII-2026 Reproducibility Package

Year: 2026

Type: software

Abstract: Reproducibility package for: Explainable Variational Quantum Policy Gradient: SHAP-Based Interpretability of Quantum Reinforcement Learning Agents on CartPole. AII 2026 Track 1, Springer CCIS. First application of SHAP to a variational quantum RL policy. VQPG achieves perfect CartPole score (500.0+/-0.0, 46 params, 200x fewer than DQN). Pole Angular Velocity dominates SHAP attributions (7x next feature).

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.21334137

Authors: Gupta, Sandeep, Hossain, Sazzad, Akhatov, Akmal, Mekhriddin, Nurmamatov, Ray, Kanad

Language: English

Topic: Explainable Artificial Intelligence (XAI)

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7168164945

Cites:

Cited by: 1

Related to: 10