The Future is Log-Gaussian: ResNets and Their Infinite-Depth-and-Width Limit at Initialization

• URL: http://arxiv.org/abs/2106.04013v1
• Date: Mon, 7 Jun 2021 23:47:37 GMT
• Title: The Future is Log-Gaussian: ResNets and Their Infinite-Depth-and-Width Limit at Initialization
• Authors: Mufan Bill Li, Mihai Nica, Daniel M. Roy
• Abstract summary: We study ReLU ResNets in the infinite-depth-and-width limit, where both depth and width tend to infinity as their ratio, $d/n$, remains constant. Using Monte Carlo simulations, we demonstrate that even basic properties of standard ResNet architectures are poorly captured by the Gaussian limit.
• Score: 18.613475245655806
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Theoretical results show that neural networks can be approximated by Gaussian processes in the infinite-width limit. However, for fully connected networks, it has been previously shown that for any fixed network width, $n$, the Gaussian approximation gets worse as the network depth, $d$, increases. Given that modern networks are deep, this raises the question of how well modern architectures, like ResNets, are captured by the infinite-width limit. To provide a better approximation, we study ReLU ResNets in the infinite-depth-and-width limit, where both depth and width tend to infinity as their ratio, $d/n$, remains constant. In contrast to the Gaussian infinite-width limit, we show theoretically that the network exhibits log-Gaussian behaviour at initialization in the infinite-depth-and-width limit, with parameters depending on the ratio $d/n$. Using Monte Carlo simulations, we demonstrate that even basic properties of standard ResNet architectures are poorly captured by the Gaussian limit, but remarkably well captured by our log-Gaussian limit. Moreover, our analysis reveals that ReLU ResNets at initialization are hypoactivated: fewer than half of the ReLUs are activated. Additionally, we calculate the interlayer correlations, which have the effect of exponentially increasing the variance of the network output. Based on our analysis, we introduce Balanced ResNets, a simple architecture modification, which eliminates hypoactivation and interlayer correlations and is more amenable to theoretical analysis.

Related papers

• Covering Numbers for Deep ReLU Networks with Applications to Function Approximation and Nonparametric Regression [4.297070083645049]
  We develop tight (up to a multiplicative constant) lower and upper bounds on the covering numbers of fully-connected networks. Thanks to the tightness of the bounds, a fundamental understanding of the impact of sparsity, quantization, bounded vs. unbounded weights, and network output truncation can be developed.
  arXiv  Detail & Related papers  (2024-10-08T21:23:14Z)
• Generalization of Scaled Deep ResNets in the Mean-Field Regime [55.77054255101667]
  We investigate emphscaled ResNet in the limit of infinitely deep and wide neural networks. Our results offer new insights into the generalization ability of deep ResNet beyond the lazy training regime.
  arXiv  Detail & Related papers  (2024-03-14T21:48:00Z)
• Wide Deep Neural Networks with Gaussian Weights are Very Close to Gaussian Processes [1.0878040851638]
  We show that the distance between the network output and the corresponding Gaussian approximation scales inversely with the width of the network, exhibiting faster convergence than the naive suggested by the central limit theorem. We also apply our bounds to obtain theoretical approximations for the exact posterior distribution of the network, when the likelihood is a bounded Lipschitz function of the network output evaluated on a (finite) training set.
  arXiv  Detail & Related papers  (2023-12-18T22:29:40Z)
• Optimization Guarantees of Unfolded ISTA and ADMM Networks With Smooth Soft-Thresholding [57.71603937699949]
  We study optimization guarantees, i.e., achieving near-zero training loss with the increase in the number of learning epochs. We show that the threshold on the number of training samples increases with the increase in the network width.
  arXiv  Detail & Related papers  (2023-09-12T13:03:47Z)
• The Onset of Variance-Limited Behavior for Networks in the Lazy and Rich Regimes [75.59720049837459]
  We study the transition from infinite-width behavior to this variance limited regime as a function of sample size $P$ and network width $N$. We find that finite-size effects can become relevant for very small datasets on the order of $P* sim sqrtN$ for regression with ReLU networks.
  arXiv  Detail & Related papers  (2022-12-23T04:48:04Z)
• On the Neural Tangent Kernel Analysis of Randomly Pruned Neural Networks [91.3755431537592]
  We study how random pruning of the weights affects a neural network's neural kernel (NTK) In particular, this work establishes an equivalence of the NTKs between a fully-connected neural network and its randomly pruned version.
  arXiv  Detail & Related papers  (2022-03-27T15:22:19Z)
• Overparameterization of deep ResNet: zero loss and mean-field analysis [19.45069138853531]
  Finding parameters in a deep neural network (NN) that fit data is a non optimization problem. We show that a basic first-order optimization method (gradient descent) finds a global solution with perfect fit in many practical situations. We give estimates of the depth and width needed to reduce the loss below a given threshold, with high probability.
  arXiv  Detail & Related papers  (2021-05-30T02:46:09Z)
• A Convergence Theory Towards Practical Over-parameterized Deep Neural Networks [56.084798078072396]
  We take a step towards closing the gap between theory and practice by significantly improving the known theoretical bounds on both the network width and the convergence time. We show that convergence to a global minimum is guaranteed for networks with quadratic widths in the sample size and linear in their depth at a time logarithmic in both. Our analysis and convergence bounds are derived via the construction of a surrogate network with fixed activation patterns that can be transformed at any time to an equivalent ReLU network of a reasonable size.
  arXiv  Detail & Related papers  (2021-01-12T00:40:45Z)
• Doubly infinite residual neural networks: a diffusion process approach [8.642603456626393]
  We show that deep ResNets do not suffer from undesirable forward-propagation properties. We focus on doubly infinite fully-connected ResNets, for which we consider i.i.d. Our results highlight a limited expressive power of doubly infinite ResNets when the unscaled network's parameters are i.i.d. and the residual blocks are shallow.
  arXiv  Detail & Related papers  (2020-07-07T07:45:34Z)
• On Random Kernels of Residual Architectures [93.94469470368988]
  We derive finite width and depth corrections for the Neural Tangent Kernel (NTK) of ResNets and DenseNets. Our findings show that in ResNets, convergence to the NTK may occur when depth and width simultaneously tend to infinity. In DenseNets, however, convergence of the NTK to its limit as the width tends to infinity is guaranteed.
  arXiv  Detail & Related papers  (2020-01-28T16:47:53Z)

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.