Faster Machine Unlearning via Natural Gradient Descent

• URL: http://arxiv.org/abs/2407.08169v1
• Date: Thu, 11 Jul 2024 04:19:28 GMT
• Title: Faster Machine Unlearning via Natural Gradient Descent
• Authors: Omri Lev, Ashia Wilson,
• Abstract summary: We address the challenge of efficiently deleting data from machine learning models using Empirical Risk Minimization (ERM), a process as machine unlearning. To avoid scratch, we propose a novel leveraging Natural Gradient Descent (D) algorithms.
• Score: 2.3020018305241337
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We address the challenge of efficiently and reliably deleting data from machine learning models trained using Empirical Risk Minimization (ERM), a process known as machine unlearning. To avoid retraining models from scratch, we propose a novel algorithm leveraging Natural Gradient Descent (NGD). Our theoretical framework ensures strong privacy guarantees for convex models, while a practical Min/Max optimization algorithm is developed for non-convex models. Comprehensive evaluations show significant improvements in privacy, computational efficiency, and generalization compared to state-of-the-art methods, advancing both the theoretical and practical aspects of machine unlearning.

Related papers

• PETScML: Second-order solvers for training regression problems in Scientific Machine Learning [0.22499166814992438]
  In recent years, we have witnessed the emergence of scientific machine learning as a data-driven tool for the analysis. We introduce a software built on top of the Portable and Extensible Toolkit for Scientific computation to bridge the gap between deep-learning software and conventional machine-learning techniques.
  arXiv  Detail & Related papers  (2024-03-18T18:59:42Z)
• Machine Unlearning of Pre-trained Large Language Models [17.40601262379265]
  This study investigates the concept of the right to be forgotten' within the context of large language models (LLMs) We explore machine unlearning as a pivotal solution, with a focus on pre-trained models.
  arXiv  Detail & Related papers  (2024-02-23T07:43:26Z)
• Machine Learning Insides OptVerse AI Solver: Design Principles and Applications [74.67495900436728]
  We present a comprehensive study on the integration of machine learning (ML) techniques into Huawei Cloud's OptVerse AI solver. We showcase our methods for generating complex SAT and MILP instances utilizing generative models that mirror multifaceted structures of real-world problem. We detail the incorporation of state-of-the-art parameter tuning algorithms which markedly elevate solver performance.
  arXiv  Detail & Related papers  (2024-01-11T15:02:15Z)
• Machine unlearning through fine-grained model parameters perturbation [26.653596302257057]
  We propose fine-grained Top-K and Random-k parameters perturbed inexact machine unlearning strategies. We also tackle the challenge of evaluating the effectiveness of machine unlearning.
  arXiv  Detail & Related papers  (2024-01-09T07:14:45Z)
• When to Update Your Model: Constrained Model-based Reinforcement Learning [50.74369835934703]
  We propose a novel and general theoretical scheme for a non-decreasing performance guarantee of model-based RL (MBRL) Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. A further example demonstrates that learning models from a dynamically-varying number of explorations benefit the eventual returns.
  arXiv  Detail & Related papers  (2022-10-15T17:57:43Z)
• Algorithms that Approximate Data Removal: New Results and Limitations [2.6905021039717987]
  We study the problem of deleting user data from machine learning models trained using empirical risk minimization. We develop an online unlearning algorithm that is both computationally and memory efficient.
  arXiv  Detail & Related papers  (2022-09-25T17:20:33Z)
• Making Linear MDPs Practical via Contrastive Representation Learning [101.75885788118131]
  It is common to address the curse of dimensionality in Markov decision processes (MDPs) by exploiting low-rank representations. We consider an alternative definition of linear MDPs that automatically ensures normalization while allowing efficient representation learning. We demonstrate superior performance over existing state-of-the-art model-based and model-free algorithms on several benchmarks.
  arXiv  Detail & Related papers  (2022-07-14T18:18:02Z)
• Large Scale Mask Optimization Via Convolutional Fourier Neural Operator and Litho-Guided Self Training [54.16367467777526]
  We present a Convolutional Neural Operator (CFCF) that can efficiently learn mask tasks. For the first time, our machine learning-based framework outperforms state-of-the-art numerical mask dataset.
  arXiv  Detail & Related papers  (2022-07-08T16:39:31Z)
• MACE: An Efficient Model-Agnostic Framework for Counterfactual Explanation [132.77005365032468]
  We propose a novel framework of Model-Agnostic Counterfactual Explanation (MACE) In our MACE approach, we propose a novel RL-based method for finding good counterfactual examples and a gradient-less descent method for improving proximity. Experiments on public datasets validate the effectiveness with better validity, sparsity and proximity.
  arXiv  Detail & Related papers  (2022-05-31T04:57:06Z)
• Physics-informed linear regression is a competitive approach compared to Machine Learning methods in building MPC [0.8135412538980287]
  We show that control in general leads to satisfactory reductions in heating and cooling energy compared to the building's baseline controller. We also see that the physics-informed ARMAX models have a lower computational burden, and a superior sample efficiency compared to the Machine Learning based models.
  arXiv  Detail & Related papers  (2021-10-29T16:56:05Z)
• Information Theoretic Model Predictive Q-Learning [64.74041985237105]
  We present a novel theoretical connection between information theoretic MPC and entropy regularized RL. We develop a Q-learning algorithm that can leverage biased models.
  arXiv  Detail & Related papers  (2019-12-31T00:29:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.