Playing Atari with Hybrid Quantum-Classical Reinforcement Learning
arXiv:2107.04114
Abstract
Despite the successes of recent works in quantum reinforcement learning, there are still severe limitations on its applications due to the challenge of encoding large observation spaces into quantum systems. To address this challenge, we propose using a neural network as a data encoder, with the Atari games as our testbed. Specifically, the neural network converts the pixel input from the games to quantum data for a Quantum Variational Circuit (QVC); this hybrid model is then used as a function approximator in the Double Deep Q Networks algorithm. We explore a number of variations of this algorithm and find that our proposed hybrid models do not achieve meaningful results on two Atari games - Breakout and Pong. We suspect this is due to the significantly reduced sizes of the hybrid quantum-classical systems.
Published in PMLR Volume 148
References in corpus (20)
- Quantum Computing in the NISQ era and beyond
- Playing Atari with Deep Reinforcement Learning
- Quantum Machine Learning
- Barren plateaus in quantum neural network training landscapes
- The theory of variational hybrid quantum-classical algorithms
- Quantum support vector machine for big data classification
- Quantum Convolutional Neural Networks
- Evaluating analytic gradients on quantum hardware
- Synthesis of Quantum Logic Circuits
- Connecting ansatz expressibility to gradient magnitudes and barren plateaus
- Quantum generative adversarial learning
- Quantum generative adversarial networks
- Quantum reinforcement learning
- Deep Reinforcement Learning that Matters
- Information-theoretic bounds on quantum advantage in machine learning
- Quantum agents in the Gym: a variational quantum algorithm for deep Q-learning
- Quantum Hopfield neural network
- Gradients of parameterized quantum gates using the parameter-shift rule and gate decomposition
- The dilemma of quantum neural networks
- The Inductive Bias of Quantum Kernels