• Corpus ID: 248810842

A hybrid classical-quantum approach to speed-up Q-learning

  title={A hybrid classical-quantum approach to speed-up Q-learning},
  author={Antonello Sannia and Alessandro Giordano and N. Lo Gullo and Carlo Mastroianni and Francesco Plastina},
We introduce a classical-quantum hybrid approach to computation, allowing for a quadratic performance improvement in the decision process of a learning agent. In particular, a quantum routine is described, which encodes on a quantum register the probability distributions that drive action choices in a reinforcement learning set-up. This routine can be employed by itself in several other contexts where decisions are driven by probabilities. After introducing the algorithm and formally evaluating… 

Figures from this paper



Speeding-up the decision making of a learning agent using an ion trap quantum processor

It is shown that the deliberation time of this quantum learning agent is quadratically improved with respect to comparable classical learning agents, highlighting the potential of scalable quantum processors taking advantage of machine learning.

Experimental quantum speed-up in reinforcement learning agents

This work presents a reinforcement learning experiment where the learning process of an agent is sped up by utilizing a quantum communication channel with the environment, and shows that combining this scenario with classical communication enables the evaluation of such an improvement, and additionally allows for optimal control of the learning progress.

Quantum agents in the Gym: a variational quantum algorithm for deep Q-learning

A training method for parametrized quantum circuits (PQCs) that can be used to solve RL tasks for discrete and continuous state spaces based on the deep Q-learning algorithm and shows when recent separation results between classical and quantum agents for policy gradient RL can be extended to inferring optimal Q-values in restricted families of environments.

Variational quantum compiling with double Q-learning

A variational quantum compiling (VQC) algorithm based on reinforcement learning is proposed in order to automatically design the structure of quantum circuit for VQC with no human intervention, and can reduce the errors of quantum algorithms due to decoherence process and gate noise in NISQ devices, and enable quantum algorithms especially for complex algorithms to be executed within coherence time.

Quantum speedup for active learning agents

It is shown that quantum physics can help and provide a quadratic speedup for active learning as a genuine problem of artificial intelligence and will be particularly relevant for applications involving complex task environments.

Variational quantum policies for reinforcement learning

This work investigates how to construct and train reinforcement learning policies based on variational quantum circuits, and proposes and shows the existence of task environments with a provable separation in performance between quantum learning agents and any polynomial-time classical learner.

Demonstration of a measurement-based adaptation protocol with quantum reinforcement learning on the IBM Q experience platform

This work tries to clone an unknown state in IBM’s QASM simulator using a quantum reinforcement learning protocol, where the “right” amount of punishment/reward function and boundary conditions can give much better fidelity than what tomography can offer in limited copies of the state.

Measurement-based adaptation protocol with quantum reinforcement learning

The experiment in this few-qubit superconducting chip faithfully reproduces the theoretical proposal, setting the first steps towards a semiautonomous quantum agent and paves the way towards quantum reinforcement learning with superconductor circuits.

Variational Quantum Circuits for Deep Reinforcement Learning

This work reshapes classical deep reinforcement learning algorithms like experience replay and target network into a representation of variational quantum circuits, and uses a quantum information encoding scheme to reduce the number of model parameters compared to classical neural networks.

Quantum Enhancements for Deep Reinforcement Learning in Large Spaces

This work studies the state-of-the-art neural-network approaches for reinforcement learning with quantum enhancements in mind and demonstrates the substantial learning advantage that models with a sampling bottleneck can provide over conventional neural network architectures in complex learning environments.