Effective Regularization Through Loss-Function Metalearning

• URL: http://arxiv.org/abs/2010.00788v2
• Date: Thu, 28 Oct 2021 04:47:05 GMT
• Title: Effective Regularization Through Loss-Function Metalearning
• Authors: Santiago Gonzalez and Risto Miikkulainen
• Abstract summary: We show that loss functions evolved by TaylorGLO balance the pull to zero error, and a push away from it to avoid overfitting. Loss-function evolution can thus be seen as a well-founded new aspect of metalearning in neural networks.
• Score: 16.8615211682877
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Evolutionary optimization, such as the TaylorGLO method, can be used to discover novel, customized loss functions for deep neural networks, resulting in improved performance, faster training, and improved data utilization. A likely explanation is that such functions discourage overfitting, leading to effective regularization. This paper demonstrates theoretically that this is indeed the case for TaylorGLO: Decomposition of learning rules makes it possible to characterize the training dynamics and show that the loss functions evolved by TaylorGLO balance the pull to zero error, and a push away from it to avoid overfitting. They may also automatically take advantage of label smoothing. This analysis leads to an invariant that can be utilized to make the metalearning process more efficient in practice; the mechanism also results in networks that are robust against adversarial attacks. Loss-function evolution can thus be seen as a well-founded new aspect of metalearning in neural networks.

Related papers

• Dynamical loss functions shape landscape topography and improve learning in artificial neural networks [0.9208007322096533]
  We show how to transform cross-entropy and mean squared error into dynamical loss functions. We show how they significantly improve validation accuracy for networks of varying sizes.
  arXiv  Detail & Related papers  (2024-10-14T16:27:03Z)
• On the Dynamics Under the Unhinged Loss and Beyond [104.49565602940699]
  We introduce the unhinged loss, a concise loss function, that offers more mathematical opportunities to analyze closed-form dynamics. The unhinged loss allows for considering more practical techniques, such as time-vary learning rates and feature normalization.
  arXiv  Detail & Related papers  (2023-12-13T02:11:07Z)
• Uncovering mesa-optimization algorithms in Transformers [61.06055590704677]
  Some autoregressive models can learn as an input sequence is processed, without undergoing any parameter changes, and without being explicitly trained to do so. We show that standard next-token prediction error minimization gives rise to a subsidiary learning algorithm that adjusts the model as new inputs are revealed. Our findings explain in-context learning as a product of autoregressive loss minimization and inform the design of new optimization-based Transformer layers.
  arXiv  Detail & Related papers  (2023-09-11T22:42:50Z)
• Outlier-Robust Neural Network Training: Efficient Optimization of Transformed Trimmed Loss with Variation Regularization [2.5628953713168685]
  We consider outlier-robust predictive modeling using highly-expressive neural networks. We employ (1) a transformed trimmed loss (TTL), which is a computationally feasible variant of the classical trimmed loss, and (2) a higher-order variation regularization (HOVR) of the prediction model.
  arXiv  Detail & Related papers  (2023-08-04T12:57:13Z)
• Theoretical Characterization of the Generalization Performance of Overfitted Meta-Learning [70.52689048213398]
  This paper studies the performance of overfitted meta-learning under a linear regression model with Gaussian features. We find new and interesting properties that do not exist in single-task linear regression. Our analysis suggests that benign overfitting is more significant and easier to observe when the noise and the diversity/fluctuation of the ground truth of each training task are large.
  arXiv  Detail & Related papers  (2023-04-09T20:36:13Z)
• Alternate Loss Functions for Classification and Robust Regression Can Improve the Accuracy of Artificial Neural Networks [6.452225158891343]
  This paper shows that training speed and final accuracy of neural networks can significantly depend on the loss function used to train neural networks. Two new classification loss functions that significantly improve performance on a wide variety of benchmark tasks are proposed.
  arXiv  Detail & Related papers  (2023-03-17T12:52:06Z)
• Online Loss Function Learning [13.744076477599707]
  Loss function learning aims to automate the task of designing a loss function for a machine learning model. We propose a new loss function learning technique for adaptively updating the loss function online after each update to the base model parameters.
  arXiv  Detail & Related papers  (2023-01-30T19:22:46Z)
• Adaptive Self-supervision Algorithms for Physics-informed Neural Networks [59.822151945132525]
  Physics-informed neural networks (PINNs) incorporate physical knowledge from the problem domain as a soft constraint on the loss function. We study the impact of the location of the collocation points on the trainability of these models. We propose a novel adaptive collocation scheme which progressively allocates more collocation points to areas where the model is making higher errors.
  arXiv  Detail & Related papers  (2022-07-08T18:17:06Z)
• Towards Scaling Difference Target Propagation by Learning Backprop Targets [64.90165892557776]
  Difference Target Propagation is a biologically-plausible learning algorithm with close relation with Gauss-Newton (GN) optimization. We propose a novel feedback weight training scheme that ensures both that DTP approximates BP and that layer-wise feedback weight training can be restored. We report the best performance ever achieved by DTP on CIFAR-10 and ImageNet.
  arXiv  Detail & Related papers  (2022-01-31T18:20:43Z)
• Gradient Starvation: A Learning Proclivity in Neural Networks [97.02382916372594]
  Gradient Starvation arises when cross-entropy loss is minimized by capturing only a subset of features relevant for the task. This work provides a theoretical explanation for the emergence of such feature imbalance in neural networks.
  arXiv  Detail & Related papers  (2020-11-18T18:52:08Z)
• Optimizing Loss Functions Through Multivariate Taylor Polynomial Parameterization [16.8615211682877]
  Loss functions are a type of metaknowledge that is crucial to effective training of deep neural network (DNN) architectures. This paper proposes continuous CMA-ES optimization of Taylor parameterizations. In MNIST, CIFAR-10, and SVHN benchmark tasks, TaylorGLO finds new loss functions that outperform functions previously discovered through GP.
  arXiv  Detail & Related papers  (2020-01-31T21:25:37Z)

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.