---
OA_place: repository
OA_type: green
_id: '22104'
abstract:
- lang: eng
  text: ' Modern computer systems store vast amounts of personal data, enabling advances
    in AI and ML but risking user privacy and trust. For privacy reasons, it is sometimes
    desired for an ML model to forget part of the data it was trained on. In this
    paper, we introduce a novel unlearning approach based on Forgetting Neural Networks
    (FNNs), a neuroscience-inspired architecture that explicitly encodes forgetting
    through multiplicative decay factors. While FNNs had previously been studied as
    a theoretical construct, we provide the first concrete implementation and demonstrate
    their effectiveness for targeted unlearning. We propose several variants with
    per-neuron forgetting factors, including rank-based assignments guided by activation
    levels, and evaluate them on MNIST and Fashion-MNIST benchmarks. Our method systematically
    removes information associated with forget sets while preserving performance on
    retained data. Membership inference attacks confirm the effectiveness of FNN-based
    unlearning in erasing information about the training data from the neural network.
    These results establish FNNs as a promising foundation for efficient and interpretable
    unlearning.'
acknowledgement: "This work has been supported by the European Research\r\nCouncil
  under Grant No.: ERC-2020-AdG\r\n101020093. We thank Brian Subirana for the original\r\nideation
  of the concept of forgetting neural networks\r\nand his belief in their potential."
article_processing_charge: No
arxiv: 1
author:
- first_name: Amartya
  full_name: Hatua, Amartya
  last_name: Hatua
- first_name: Trung T.
  full_name: Nguyen, Trung T.
  last_name: Nguyen
- first_name: Filip
  full_name: Cano Cordoba, Filip
  id: 708cad98-e86a-11ef-8098-bdae2d7c6af1
  last_name: Cano Cordoba
  orcid: 0000-0002-0783-904X
- first_name: Andrew H.
  full_name: Sung, Andrew H.
  last_name: Sung
citation:
  ama: 'Hatua A, Nguyen TT, Cano Cordoba F, Sung AH. Machine unlearning using forgetting
    neural networks. In: <i>Proceedings of the 18th International Conference on Agents
    and Artificial Intelligence</i>. Vol 2. Science and Technology Publications; 2026:1536-1546.
    doi:<a href="https://doi.org/10.5220/0014326500004052">10.5220/0014326500004052</a>'
  apa: 'Hatua, A., Nguyen, T. T., Cano Cordoba, F., &#38; Sung, A. H. (2026). Machine
    unlearning using forgetting neural networks. In <i>Proceedings of the 18th International
    Conference on Agents and Artificial Intelligence</i> (Vol. 2, pp. 1536–1546).
    Marbella, Spain: Science and Technology Publications. <a href="https://doi.org/10.5220/0014326500004052">https://doi.org/10.5220/0014326500004052</a>'
  chicago: Hatua, Amartya, Trung T. Nguyen, Filip Cano Cordoba, and Andrew H. Sung.
    “Machine Unlearning Using Forgetting Neural Networks.” In <i>Proceedings of the
    18th International Conference on Agents and Artificial Intelligence</i>, 2:1536–46.
    Science and Technology Publications, 2026. <a href="https://doi.org/10.5220/0014326500004052">https://doi.org/10.5220/0014326500004052</a>.
  ieee: A. Hatua, T. T. Nguyen, F. Cano Cordoba, and A. H. Sung, “Machine unlearning
    using forgetting neural networks,” in <i>Proceedings of the 18th International
    Conference on Agents and Artificial Intelligence</i>, Marbella, Spain, 2026, vol.
    2, pp. 1536–1546.
  ista: 'Hatua A, Nguyen TT, Cano Cordoba F, Sung AH. 2026. Machine unlearning using
    forgetting neural networks. Proceedings of the 18th International Conference on
    Agents and Artificial Intelligence. ICAART: International Conference on Agents
    and Artificial Intelligence vol. 2, 1536–1546.'
  mla: Hatua, Amartya, et al. “Machine Unlearning Using Forgetting Neural Networks.”
    <i>Proceedings of the 18th International Conference on Agents and Artificial Intelligence</i>,
    vol. 2, Science and Technology Publications, 2026, pp. 1536–46, doi:<a href="https://doi.org/10.5220/0014326500004052">10.5220/0014326500004052</a>.
  short: A. Hatua, T.T. Nguyen, F. Cano Cordoba, A.H. Sung, in:, Proceedings of the
    18th International Conference on Agents and Artificial Intelligence, Science and
    Technology Publications, 2026, pp. 1536–1546.
conference:
  end_date: 2026-03-08
  location: Marbella, Spain
  name: 'ICAART: International Conference on Agents and Artificial Intelligence'
  start_date: 2026-03-05
das_tickbox: '0'
date_created: 2026-06-21T22:03:01Z
date_published: 2026-04-01T00:00:00Z
date_updated: 2026-10-20T14:05:02Z
day: '01'
ddc:
- '000'
department:
- _id: ToHe
doi: 10.5220/0014326500004052
ec_funded: 1
external_id:
  arxiv:
  - '2410.22374'
file:
- access_level: open_access
  checksum: d0f5212846c8f3129509b081648465cb
  content_type: application/pdf
  creator: dernst
  date_created: 2026-10-20T14:00:27Z
  date_updated: 2026-10-20T14:00:27Z
  file_id: '23178'
  file_name: 2026_CAART_Hatua.pdf
  file_size: 779692
  relation: main_file
  success: 1
file_date_updated: 2026-10-20T14:00:27Z
fulldoi: https://doi.org/10.5220/0014326500004052
has_accepted_license: '1'
intvolume: '         2'
keyword:
- Machine Unlearning
- Neuroscience-Inspired Machine Learning
- Membership Inference Attacks
language:
- iso: eng
month: '04'
oa: 1
oa_version: Published Version
page: 1536-1546
project:
- _id: 62781420-2b32-11ec-9570-8d9b63373d4d
  call_identifier: H2020
  grant_number: '101020093'
  name: Vigilant Algorithmic Monitoring of Software
publication: Proceedings of the 18th International Conference on Agents and Artificial
  Intelligence
publication_identifier:
  eissn:
  - 2184-433X
  isbn:
  - '9789897587962'
  issn:
  - 2184-3589
publication_status: published
publisher: Science and Technology Publications
quality_controlled: '1'
researchdata_availability: no
scopus_import: '1'
status: public
supplementarymaterial: no
title: Machine unlearning using forgetting neural networks
tmp:
  image: /images/cc_by_nc_nd.png
  legal_code_url: https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode
  name: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
    (CC BY-NC-ND 4.0)
  short: CC BY-NC-ND (4.0)
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 2
year: '2026'
...