Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Reset

• URL: http://arxiv.org/abs/2411.04034v1
• Date: Wed, 06 Nov 2024 16:32:40 GMT
• Title: Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Reset
• Authors: Alexandre Galashov, Michalis K. Titsias, András György, Clare Lyle, Razvan Pascanu, Yee Whye Teh, Maneesh Sahani,
• Abstract summary: We introduce a novel learning approach that automatically models and adapts to non-stationarity. We show empirically that our approach performs well in non-stationary supervised and off-policy reinforcement learning settings.
• Score: 98.52916361979503
• License:
• Abstract: Neural networks are traditionally trained under the assumption that data come from a stationary distribution. However, settings which violate this assumption are becoming more popular; examples include supervised learning under distributional shifts, reinforcement learning, continual learning and non-stationary contextual bandits. In this work we introduce a novel learning approach that automatically models and adapts to non-stationarity, via an Ornstein-Uhlenbeck process with an adaptive drift parameter. The adaptive drift tends to draw the parameters towards the initialisation distribution, so the approach can be understood as a form of soft parameter reset. We show empirically that our approach performs well in non-stationary supervised and off-policy reinforcement learning settings.

Related papers

• Dreaming Learning [41.94295877935867]
  Introducing new information to a machine learning system can interfere with previously stored data. We propose a training algorithm inspired by Stuart Kauffman's notion of the Adjacent Possible. It predisposes the neural network to smoothly accept and integrate data sequences with different statistical characteristics than expected.
  arXiv  Detail & Related papers  (2024-10-23T09:17:31Z)
• Liquid Neural Network-based Adaptive Learning vs. Incremental Learning for Link Load Prediction amid Concept Drift due to Network Failures [37.66676003679306]
  Adapting to concept drift is a challenging task in machine learning. In communication networks, such issue emerges when performing traffic forecasting following afailure event. We propose an approach that exploits adaptive learning algorithms, namely, liquid neural networks, which are capable of self-adaptation to abrupt changes in data patterns without requiring any retraining.
  arXiv  Detail & Related papers  (2024-04-08T08:47:46Z)
• Federated Learning with Projected Trajectory Regularization [65.6266768678291]
  Federated learning enables joint training of machine learning models from distributed clients without sharing their local data. One key challenge in federated learning is to handle non-identically distributed data across the clients. We propose a novel federated learning framework with projected trajectory regularization (FedPTR) for tackling the data issue.
  arXiv  Detail & Related papers  (2023-12-22T02:12:08Z)
• Kalman Filter for Online Classification of Non-Stationary Data [101.26838049872651]
  In Online Continual Learning (OCL) a learning system receives a stream of data and sequentially performs prediction and training steps. We introduce a probabilistic Bayesian online learning model by using a neural representation and a state space model over the linear predictor weights. In experiments in multi-class classification we demonstrate the predictive ability of the model and its flexibility to capture non-stationarity.
  arXiv  Detail & Related papers  (2023-06-14T11:41:42Z)
• Unsupervised Unlearning of Concept Drift with Autoencoders [5.41354952642957]
  Concept drift refers to a change in the data distribution affecting the data stream of future samples. This paper proposes an unsupervised and model-agnostic concept drift adaptation method at the global level.
  arXiv  Detail & Related papers  (2022-11-23T14:52:49Z)
• On the Forward Invariance of Neural ODEs [92.07281135902922]
  We propose a new method to ensure neural ordinary differential equations (ODEs) satisfy output specifications. Our approach uses a class of control barrier functions to transform output specifications into constraints on the parameters and inputs of the learning system.
  arXiv  Detail & Related papers  (2022-10-10T15:18:28Z)
• Learning to Learn with Generative Models of Neural Network Checkpoints [71.06722933442956]
  We construct a dataset of neural network checkpoints and train a generative model on the parameters. We find that our approach successfully generates parameters for a wide range of loss prompts. We apply our method to different neural network architectures and tasks in supervised and reinforcement learning.
  arXiv  Detail & Related papers  (2022-09-26T17:59:58Z)
• Recurrent Neural Network Training with Convex Loss and Regularization Functions by Extended Kalman Filtering [0.20305676256390928]
  We show that the learning method outperforms gradient descent in a nonlinear system identification benchmark. We also explore the use of the algorithm in data-driven nonlinear model predictive control and its relation with disturbance models for offset-free tracking.
  arXiv  Detail & Related papers  (2021-11-04T07:49:15Z)
• Automatic Recall Machines: Internal Replay, Continual Learning and the Brain [104.38824285741248]
  Replay in neural networks involves training on sequential data with memorized samples, which counteracts forgetting of previous behavior caused by non-stationarity. We present a method where these auxiliary samples are generated on the fly, given only the model that is being trained for the assessed objective. Instead the implicit memory of learned samples within the assessed model itself is exploited.
  arXiv  Detail & Related papers  (2020-06-22T15:07:06Z)
• Technical Report: Adaptive Control for Linearizable Systems Using On-Policy Reinforcement Learning [41.24484153212002]
  This paper proposes a framework for adaptively learning a feedback linearization-based tracking controller for an unknown system. It does not require the learned inverse model to be invertible at all instances of time. A simulated example of a double pendulum demonstrates the utility of the proposed theory.
  arXiv  Detail & Related papers  (2020-04-06T15:50:31Z)

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.