Q Learning with Quantum Neural Networks

Hu, Wei; Hu, James Lee

doi:10.4236/ns.2019.111005

Cited by 6 publications

(7 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our work shows how to implement Q learning and actor-critic algorithms using deep quantum neural networks. The results of this report highlight that having more layers in the quantum networks does improve the learning of the agent in the grid world environment, which extends our previous work that uses a single layer of quantum network to solve RL problems [21] [22].…”

Section: Discussionsupporting

confidence: 72%

Reinforcement Learning with Deep Quantum Neural Networks

Hu¹,

Hu²

2019

JQIS

Self Cite

View full text Add to dashboard Cite

The advantage of quantum computers over classical computers fuels the recent trend of developing machine learning algorithms on quantum computers, which can potentially lead to breakthroughs and new learning models in this area. The aim of our study is to explore deep quantum reinforcement learning (RL) on photonic quantum computers, which can process information stored in the quantum states of light. These quantum computers can naturally represent continuous variables, making them an ideal platform to create quantum versions of neural networks. Using quantum photonic circuits, we implement Q learning and actor-critic algorithms with multilayer quantum neural networks and test them in the grid world environment. Our experiments show that 1) these quantum algorithms can solve the RL problem and 2) compared to one layer, using three layer quantum networks improves the learning of both algorithms in terms of rewards collected. In summary, our findings suggest that having more layers in deep quantum RL can enhance the learning outcome.

show abstract

Section: Discussionsupporting

confidence: 72%

Reinforcement Learning with Deep Quantum Neural Networks

Hu¹,

Hu²

2019

JQIS

Self Cite

View full text Add to dashboard Cite

show abstract

“…Despite advances in mass spectrometry instrumentation, [1][2][3][4] fragmentation techniques, [5][6][7][8][9] and data analysis software, [10][11][12] intact protein (or top-down) analysis suffers from severe analytical limitations compared to peptide mass analysis. Higher sequence coverage across more proteoforms of larger molecular weight (MW) is required to eliminate current analytical "blind-spots" associated with the approach.…”

Section: Introductionmentioning

confidence: 99%

Increased Single-Spectrum Top-Down Protein Sequence Coverage in Trapping Mass Spectrometers with Chimeric Ion Loading

et al. 2020

View full text Add to dashboard Cite

Fourier transform mass spectrometers routinely provide high mass resolution, mass measurement accuracy, and mass spectral dynamic range. In this work, we utilize 21 T Fourier transform ion cyclotron resonance (FT-ICR) to analyze product ions derived from the application of multiple dissociation techniques and/or multiple precursor ions within a single transient acquisition. This ion loading technique, which we call, "chimeric ion loading", saves valuable acquisition time, decreases sample consumption, and improves top-down protein sequence coverage. In the analysis of MCF7 cell lysate, we show collision-induced dissociation (CID) and electron-transfer dissociation (ETD) on each precursor on a liquid chromatography-mass spectrometry (LC-MS) timescale and improve mean sequence coverage dramatically (CID-only 15% vs chimeric 33%),

show abstract

“…a the parameter counts are denoted for a single agent; Q-Learning with Quantum Neural Networks, Summary. The paper by Hu and Hu [HH19a] uses a QNN to approximate the action-value function in Q-learning-based RL. The proposed algorithm is tested in the discrete state-action space FrozenLake environment (of size 2 × 3) from OpenAI Gym.…”

Section: Algorithmicmentioning

confidence: 99%

A Survey on Quantum Reinforcement Learning

Meyer¹,

Ufrecht²,

Periyasamy³

et al. 2022

Preprint

View full text Add to dashboard Cite

Quantum reinforcement learning is an emerging field at the intersection of quantum computing and machine learning. While we intend to provide a broad overview of the literature on quantum reinforcement learning (our interpretation of this term will be clarified below), we put particular emphasis on recent developments. With a focus on already available noisy intermediate-scale quantum devices, these include variational quantum circuits acting as function approximators in an otherwise classical reinforcement learning setting. In addition, we survey quantum reinforcement learning algorithms based on future fault-tolerant hardware, some of which come with a provable quantum advantage. We provide both a birds-eye-view of the field, as well as summaries and reviews for selected parts of the literature.

show abstract

Q Learning with Quantum Neural Networks

Cited by 6 publications

References 23 publications

Reinforcement Learning with Deep Quantum Neural Networks

Reinforcement Learning with Deep Quantum Neural Networks

Increased Single-Spectrum Top-Down Protein Sequence Coverage in Trapping Mass Spectrometers with Chimeric Ion Loading

A Survey on Quantum Reinforcement Learning

Contact Info

Product

Resources

About