Data

Tools

Indexes

Applications

Experiments

Open Source Science

learning-complexity-gradually

Repository of the paper "Learning complexity gradually in quantum machine learning models"

https://github.com/erikrecio/learning-complexity-gradually

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (9.9%) to scientific vocabulary

Keywords

classification curriculum-learning inductive-bias quantum-computing quantum-machine-learning
Last synced: 4 months ago · JSON representation ·

Repository

Repository of the paper "Learning complexity gradually in quantum machine learning models"

Basic Info
Statistics
  • Stars: 1
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
classification curriculum-learning inductive-bias quantum-computing quantum-machine-learning
Created about 1 year ago · Last pushed 8 months ago
Metadata Files
Readme Citation

README.md

Learning complexity gradually in quantum machine learning models

This repository contains the functions and scripts used for the simulation and optimization of quantum neural networks supporting the research presented in the paper "Learning complexity gradually in quantum machine learning models", available on arxiv. Dependencies are listed in requirements.txt.

Repository structure and running the code

This repository contains various utility files and three main executable scripts:

  1. main.py: This is the core script where training and optimization occur. Before running, adjust the hyperparameters at the start of the script to suit your needs. When executed, this script generates multiple CSV files in the Results folder, containing relevant training data, and creates loss plots to visualize the progress of each training strategy.

  2. plot_accuracy.py: This script generates a figure (saved in the Results folder) that compares the accuracies across different training strategies. Customize the parameters within the script to specify the target files, then run the script to produce the accuracy comparison figure.

  3. plot_probabilities.py: This script outputs a figure illustrating the classification probabilities for different classes along a path in the parametrized Hamiltonian space. Adjust the settings as needed to obtain the desired visualization. The output is also saved in the Results folder.

In the Results folder, you’ll find examples of the outputs generated by these scripts. Note that, due to size constraints, the larger CSV files containing raw data have not been included in this repository.

Owner

  • Login: erikrecio
  • Kind: user

Citation (CITATION.cff) 

cff-version: 1.2.0
title: Learning complexity gradually in QML models repository
message: >-
  If you use this software, please cite it using the
  metadata from this file.
type: software
authors:
  - given-names: Erik
    family-names: Recio-Armengol
    email: erik.recio@icfo.eu
    affiliation: ICFO (Institute of Photonic Sciences)
  - given-names: Carlos
    family-names: Bravo-Prieto
    email: c.bravo.prieto@fu-berlin.de
    affiliation: FU Berlin
identifiers:
  - type: url
    value: https://arxiv.org/abs/2411.11954
    description: Paper's arxiv
repository-code: https://github.com/erikrecio/learning-complexity-gradually
keywords:
  - Curriculum Learning
  - Inductive Bias
  - Quantum Machine Learning
preferred-citation:
  type: generic
  title: Learning complexity gradually in quantum machine learning models
  authors:
  - given-names: Erik
    family-names: Recio-Armengol
    affiliation: ICFO (Institute of Photonic Sciences)
  - given-names: Franz J.
    family-names: Schreiber
    affiliation: FU Berlin
  - given-names: Jens
    family-names: Eisert
    affiliation: FU Berlin
  - given-names: Carlos
    family-names: Bravo-Prieto
    affiliation: FU Berlin
  year: 2024
  eprint: 2411.11954
  archivePrefix: arXiv
  primaryClass: quant-ph
  url: https://arxiv.org/abs/2411.11954

GitHub Events

Total
  • Push event: 11
Last Year
  • Push event: 11

Dependencies

requirements.txt pypi
  • PennyLane ==0.37.0
  • SciencePlots ==2.1.1
  • Shapely ==2.0.6
  • jax ==0.4.31
  • jaxopt ==0.8.3
  • latex ==0.7.0
  • matplotlib ==3.9.2
  • numpy ==2.1.3
  • optax ==0.2.3
  • pandas ==2.2.3
  • pypdf ==5.1.0
  • pytz ==2024.1