skip to main content

This content will become publicly available on July 24, 2024

Title: Function-Space Regularization in Neural Networks
Parameter-space regularization in neural network optimization is a fundamental tool for improving generalization. However, standard parameter-space regularization methods make it challenging to encode explicit preferences about desired predictive functions into neural network training. In this work, we approach regularization in neural networks from a probabilistic perspective and show that by viewing parameter-space regularization as specifying an empirical prior distribution over the model parameters, we can derive a probabilistically well-motivated regularization technique that allows explicitly encoding information about desired predictive functions into neural network training. This method—which we refer to as function-space empirical Bayes (FS-EB)—includes both parameter- and function-space regularization, is mathematically simple, easy to implement, and incurs only minimal computational overhead compared to standard regularization techniques. We evaluate the utility of this regularization technique empirically and demonstrate that the proposed method leads to near-perfect semantic shift detection, highly-calibrated predictive uncertainty estimates, successful task adaption from pre-trained models, and improved generalization under covariate shift.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ; ;
Publisher / Repository:
International Conference on Machine Learning
Date Published:
Journal Name:
International Conference on Machine Learning
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Mixup is a popular regularization technique for training deep neural networks that improves generalization and increases robustness to certain distribution shifts. It perturbs input training data in the direction of other randomly-chosen instances in the training set. To better leverage the structure of the data, we extend mixup in a simple, broadly applicable way to k-mixup, which perturbs k-batches of training points in the direction of other k-batches. The perturbation is done with displacement interpolation, i.e. interpolation under the Wasserstein metric. We demonstrate theoretically and in simulations that k-mixup preserves cluster and manifold structures, and we extend theory studying the efficacy of standard mixup to the k-mixup case. Our empirical results show that training with k-mixup further improves generalization and robustness across several network architectures and benchmark datasets of differing modalities. For the wide variety of real datasets considered, the performance gains of k-mixup over standard mixup are similar to or larger than the gains of mixup itself over standard ERM after hyperparameter optimization. In several instances, in fact, k-mixup achieves gains in settings where standard mixup has negligible to zero improvement over ERM. 
    more » « less
  2. Rich variations in gait are generated according to several attributes of the individual and environment, such as age, athleticism, terrain, speed, personal “style”, mood, etc. The effects of these attributes can be hard to quantify explicitly, but relatively straightforward to sample. We seek to generate gait that expresses these attributes, creating synthetic gait samples that exemplify a custom mix of attributes. This is difficult to perform manually, and generally restricted to simple, human-interpretable and handcrafted rules. In this manuscript, we present neural network architectures to learn representations of hard to quantify attributes from data, and generate gait trajectories by composing multiple desirable attributes. We demonstrate this method for the two most commonly desired attribute classes: individual style and walking speed. We show that two methods, cost function design and latent space regularization, can be used individually or combined. We also show two uses of machine learning classifiers that recognize individuals and speeds. Firstly, they can be used as quantitative measures of success; if a synthetic gait fools a classifier, then it is considered to be a good example of that class. Secondly, we show that classifiers can be used in the latent space regularizations and cost functions to improve training beyond a typical squared-error cost. 
    more » « less
  3. For any given neural network architecture a permutation of weights and biases results in the same functional network. This implies that optimization algorithms used to 'train' or 'learn' the network are faced with a very large number (in the millions even for small networks) of equivalent optimal solutions in the parameter space. To the best of our knowledge, this observation is absent in the literature. In order to narrow down the parameter search space, a novel technique is introduced in order to fix the bias vector configurations to be monotonically increasing. This is achieved by augmenting a typical learning problem with inequality constraints on the bias vectors in each layer. A Moreau-Yosida regularization based algorithm is proposed to handle these inequality constraints and a theoretical convergence of this algorithm is established. Applications of the proposed approach to standard trigonometric functions and more challenging stiff ordinary differential equations arising in chemically reacting flows clearly illustrate the benefits of the proposed approach. Further application of the approach on the MNIST dataset within TensorFlow, illustrate that the presented approach can be incorporated in any of the existing machine learning libraries.

    more » « less
  4. Feature representations from pre-trained deep neural networks have been known to exhibit excellent generalization and utility across a variety of related tasks. Fine-tuning is by far the simplest and most widely used approach that seeks to exploit and adapt these feature representations to novel tasks with limited data. Despite the effectiveness of fine-tuning, itis often sub-optimal and requires very careful optimization to prevent severe over-fitting to small datasets. The problem of sub-optimality and over-fitting, is due in part to the large number of parameters used in a typical deep convolutional neural network. To address these problems, we propose a simple yet effective regularization method for fine-tuning pre-trained deep networks for the task of k-shot learning. To prevent overfitting, our key strategy is to cluster the model parameters while ensuring intra-cluster similarity and inter-cluster diversity of the parameters, effectively regularizing the dimensionality of the parameter search space. In particular, we identify groups of neurons within each layer of a deep network that shares similar activation patterns. When the network is to be fine-tuned for a classification task using only k examples, we propagate a single gradient to all of the neuron parameters that belong to the same group. The grouping of neurons is non-trivial as neuron activations depend on the distribution of the input data. To efficiently search for optimal groupings conditioned on the input data, we propose a reinforcement learning search strategy using recurrent networks to learn the optimal group assignments for each network layer. Experimental results show that our method can be easily applied to several popular convolutional neural networks and improve upon other state-of-the-art fine-tuning based k-shot learning strategies by more than10% 
    more » « less
  5. The prosperity of deep learning and automated machine learning (AutoML) is largely rooted in the development of novel neural networks -- but what defines and controls the "goodness" of networks in an architecture space? Test accuracy, a golden standard in AutoML, is closely related to three aspects: (1) expressivity (how complicated functions a network can approximate over the training data); (2) convergence (how fast the network can reach low training error under gradient descent); (3) generalization (whether a trained network can be generalized from the training data to unseen samples with low test error). However, most previous theory papers focus on fixed model structures, largely ignoring sophisticated networks used in practice. To facilitate the interpretation and understanding of the architecture design by AutoML, we target connecting a bigger picture: how does the architecture jointly impact its expressivity, convergence, and generalization? We demonstrate the "no free lunch" behavior in networks from an architecture space: given a fixed budget on the number of parameters, there does not exist a single architecture that is optimal in all three aspects. In other words, separately optimizing expressivity, convergence, and generalization will achieve different networks in the architecture space. Our analysis can explain a wide range of observations in AutoML. Experiments on popular benchmarks confirm our theoretical analysis. Our codes are attached in the supplement. 
    more » « less