New methods train neural networks without changing weights, achieving similar or higher performance.
problem Training neural networks efficiently with randomly initialized weights.
method Switching connections on and off, flipping weights' signs, minimizing changed connections.
result Achieves similar or higher performance with less computational cost than training all weights.
A new method improves RVFL networks for resource-efficient machine learning.
problem Deploying machine learning on edge devices with limited resources.
method Density encoding and hyperdimensional computing operations.
result The proposed method achieves higher accuracy and lower energy consumption.
Bounds neural network output distribution to Gaussian for random initialization.
problem Quantifying the distribution of randomly initialized deep neural networks.
method Quantitative Gaussian approximation using quadratic Wasserstein distance.
result Explicit inequalities show how network sizes affect Gaussian behavior.
Randomly initialized neural networks can linearly separate arbitrary sets.
problem Mapping two arbitrary sets to linearly separable sets.
method Randomly initialized one-layer neural networks with sufficient width.
result With high probability, these networks can transform two sets into linearly separable sets.
Batch normalization prevents rank collapse in deep networks, improving training stability.
problem Rank collapse in randomly initialized deep networks with increasing depth.
method Investigates spectral instabilities in random matrices and uses batch normalization to avoid rank collapse.
result Batch normalization prevents rank collapse in both linear and ReLU networks, improving training stability.
RicciNets prunes neural networks by removing edges of low importance based on Ricci curvature, reducing FLOPs by 35%.
problem Pruning neural networks to reduce computational load and improve efficiency.
method RicciNets uses Ricci curvature to prune edges of low importance in a randomly wired neural network, reducing FLOPs.
result Reduction of almost 35% in FLOPs with no performance degradation.
One of the mysteries in the success of neural networks is randomly initialized first order methods like gradient descent can achieve zero training loss even though the objective function is non-convex and non-smooth. This paper demystifies this surprising phenomenon for two-layer fully connected ReLU activated neural n…
The past few years have witnessed the fast development of different regularization methods for deep learning models such as fully-connected deep neural networks (DNNs) and Convolutional Neural Networks (CNNs). Most of previous methods mainly consider to drop features from input data and hidden layers, such as Dropout, …
This paper is concerned with the sparsification of the input-hidden weights of ELM (Extreme Learning Machine). For ordinary feedforward neural networks, the sparsification is usually done by introducing certain regularization technique into the learning process of the network. But this strategy can not be applied for E…
A new method to prune neural networks with iterative randomization improves efficiency.
problem Efficiency in pruning randomly initialized neural networks.
method Iteratively randomizing weight values to reduce parameter requirements.
result The method achieves remarkable performance with fewer parameters.
Paper presents a self-adaptive learning model for robust classification and regression.
problem Dealing with various datasets of different complexity.
method Combines DNDN and DSP, an end-to-end training approach with multiple randomly initialized softmax layers and adaptive soft pruning.
result The model demonstrates no performance loss compared with unpruned models and higher robustness over different data and feature distributions.
RWTNs improve NTN performance in SRL tasks.
problem Improving NTN performance in SRL tasks.
method Randomly weighted, untrained tensors in NTN encoder network.
result RWTNs outperform LTNs in SII tasks.
Explains neural network properties with a simple operation.
problem Understanding neural network performance and generalization.
method Analyze a single operation: expand-and-sparsify.
result Clarifies phenomena like Lottery Ticket Hypothesis and overparameterized models.
Study on dropout in neural networks using percolation theory.
problem Understanding dropout's effect on neural network training.
method Investigates percolation models mimicking dropout in neural networks.
result Dropout can cause a breakdown in neural networks.
Deep networks can perfectly classify two low-dimensional manifolds on a sphere with large depth and width.
problem Binary classification of two low-dimensional submanifolds on a sphere.
method Analysis of a deep fully-connected neural network trained to separate two submanifolds of the unit sphere.
result Randomly-initialized gradient descent can perfectly classify the two manifolds with high probability when the network depth is large relative to certain geometric and statistical properties of the data.
This paper investigates the role of sparsity in Reservoir Computing networks.
problem Designing efficient Recurrent Neural Networks (RNNs) with hidden recurrent layers.
method Empirical investigation of sparsity in input-reservoir connections and recurrent connections.
result Sparsity, particularly in input-reservoir connections, enhances the network's temporal memory and dimensionality.
The angular power spectrum characterizes neural network complexity.
problem Characterizing the complexity of deep neural networks.
method Using the angular power spectrum of the limiting field to characterize network complexity.
result Classified neural networks as low-disorder, sparse, or high-disorder.
Network pruning is a promising avenue for compressing deep neural networks. A typical approach to pruning starts by training a model and then removing redundant parameters while minimizing the impact on what is learned. Alternatively, a recent approach shows that pruning can be done at initialization prior to training,…
Study on the complexity of 1D ReLU neural networks, proving growth in linear regions.
problem Understanding the complexity and expressivity of 1D ReLU neural networks.
method Analyzing the number of linear regions in randomly initialized, fully connected 1D ReLU networks in the infinite-width limit.
result The expected number of linear regions grows as a function of the number of neurons in each layer.
Recent years have witnessed the success of deep neural networks in dealing with a plenty of practical problems. Dropout has played an essential role in many successful deep neural networks, by inducing regularization in the model training. In this paper, we present a new regularized training approach: Shakeout. Instead…
Neural networks are mathematically represented via quiver representations.
problem Understanding how neural networks process data and create representations.
method Representing neural networks as quiver representations with activation functions.
result Neural networks' computations can be studied algebraically and geometrically.
Randomly initialized networks contain subnetworks that perform similarly to target networks.
problem Proving the lottery ticket hypothesis for neural networks.
method Pruning over-parameterized neural networks to find subnetworks.
result Randomly initialized networks contain subnetworks with similar performance to target networks without additional training.
Randomly initialized wide neural networks with zero-mean activations are nearly independent, potentially solving AI interpretability limits.
problem Measuring the limits of AI interpretability.
method Randomly initialized neural networks with large width and zero-mean activation functions.
result Neural networks with zero-mean activations are nearly independent, solving the computational no-coincidence conjecture.
Optimizes wide low-rank neural networks for reduced parameters and cost.
problem Reducing the number of learnable parameters in wide neural networks.
method Analyzed edge-of-chaos dynamics and derived formulae for optimal weight and bias variances.
result Optimal weight and bias variances for low-rank networks follow from multiplicative scaling.
DropCluster clusters features in convolutional layers to prevent overfitting.
problem Dropout's effectiveness in convolutional layers is limited due to feature spatial correlation.
method DropCluster clusters features in convolutional layers and randomly drops clusters during training.
result DropCluster controls overfitting better than other approaches on various datasets.
DRE combines DNN with random feature regression for efficient neural network design.
problem Designing and training deep neural networks (DNN) efficiently and effectively.
method DRE architecture with two-layer neural networks, randomly drawn input and output weights trained with linear ridge regression.
result DRE outperforms state-of-the-art DNN in many data sets with lower computational cost.
Layered neural networks have greatly improved the performance of various applications including image processing, speech recognition, natural language processing, and bioinformatics. However, it is still difficult to discover or interpret knowledge from the inference provided by a layered neural network, since its inte…
Neural networks learn patterns in random data, improving downstream performance.
problem Understanding what deep networks learn with random labels.
method Analytical and empirical study of convolutional and fully connected networks pre-trained on random labels.
result Pre-trained networks on random labels transfer faster to real datasets, despite specialization effects.
In this work, a method of random parameters generation for randomized learning of a single-hidden-layer feedforward neural network is proposed. The method firstly, randomly selects the slope angles of the hidden neurons activation functions from an interval adjusted to the target function, then randomly rotates the act…
New proof links initial class bias to DNN trainability, challenging traditional understanding.
problem Understanding the initial class bias in DNNs and its impact on trainability.
method Theoretical proof linking initial class bias to mean field theories of DNNs.
result Efficient learning is connected to a network's prejudice towards a specific class, contradicting traditional understanding.
We give a rigorous analysis of the statistical behavior of gradients in a randomly initialized fully connected network N with ReLU activations. Our results show that the empirical variance of the squares of the entries in the input-output Jacobian of N is exponential in a simple architecture-dependent constant beta, gi…
Deep learning has become a powerful and popular tool for a variety of machine learning tasks. However, it is challenging to understand the mechanism of deep learning from a theoretical perspective. In this work, we propose a random active path model to study collective properties of deep neural networks with binary syn…
Dropout is used to avoid overfitting by randomly dropping units from the neural networks during training. Inspired by dropout, this paper presents GI-Dropout, a novel dropout method integrating with global information to improve neural networks for text classification. Unlike the traditional dropout method in which the…
New reservoir computing approach handles infinite-dimensional systems.
problem Approximating and generalizing complex input/output systems.
method Randomly generated echo state networks with neural networks.
result Proves universal approximation properties for new class of systems.
Dropout is a simple but efficient regularization technique for achieving better generalization of deep neural networks (DNNs); hence it is widely used in tasks based on DNNs. During training, dropout randomly discards a portion of the neurons to avoid overfitting. This paper presents an enhanced dropout technique, whic…
Randomly chosen primary hidden units and derived secondary units reduce neural network complexity.
problem Large number of hidden units in neural networks.
method Introducing primary and secondary hidden units with random weights for primary units and derived weights for secondary units.
result Significant reduction in the number of hidden units without compromising accuracy.
New network learns non-parametric invariances from data.
problem Modeling non-parametric invariances in data.
method Introduces PRC-NPTN networks with permanent random connectomes.
result Improves generalization and outperforms existing methods.
Almost all local minima in neural networks are strongly convex.
problem The prevalence of strongly convex neighborhoods around local minima in neural network optimization landscapes.
method Rigorous analysis of shallow neural networks with analytic activation functions, dividing parameter space into efficient and redundant domains.
result For shallow neural networks on the efficient domain, almost all local minima are strongly convex.
Deep neural networks (DNNs) often require good regularizers to generalize well. Currently, state-of-the-art DNN regularization techniques consist in randomly dropping units and/or connections on each iteration of the training algorithm. Dropout and DropConnect are characteristic examples of such regularizers, that are …
Meta-ticket finds optimal sparse subnetworks for few-shot learning in randomly initialized neural networks.
problem Avoiding overfitting in few-shot learning for over-parameterized neural networks.
method Meta-learning approach to find optimal sparse subnetworks.
result Meta-ticket discovers sparse subnetworks that adapt to each task, achieving superior meta-generalization.
An interesting approach to analyzing neural networks that has received renewed attention is to examine the equivalent kernel of the neural network. This is based on the fact that a fully connected feedforward network with one hidden layer, a certain weight distribution, an activation function, and an infinite number of…
Deep neural networks (DNNs) have demonstrated dominating performance in many fields; since AlexNet, networks used in practice are going wider and deeper. On the theoretical side, a long line of works has been focusing on training neural networks with one hidden layer. The theory of multi-layer networks remains largely …
We generalize the scale-free network model of Barabàsi and Albert [Science 286, 509 (1999)] by proposing a class of stochastic models for scale-free interdependent networks in which interdependent nodes are not randomly connected but rather are connected via preferential attachment (PA). Each network grows through the …
This article presents a new methodology called deep Theory of Functional Connections (TFC) that estimates the solutions of partial differential equations (PDEs) by combining neural networks with TFC. TFC is used to transform PDEs with boundary conditions into unconstrained optimization problems by embedding the boundar…
Paper optimizes neural network initialization using SMT solvers.
problem Improving neural network performance through better initialization.
method Reduces initialization to SMT problem solving.
result Proposed method achieves better performance than random initialization.
Monte Carlo method trains deep neural networks without gradients.
problem Vanishing and exploding gradients in backpropagation.
method Randomly mutate parameters, keep if loss decreases.
result Gradient-free method trains deep networks effectively.
New bounds on neural network convergence using information theory.
problem Quantifying convergence rates of neural networks to Gaussian distributions.
method Entropic inequalities and Gaussian approximations.
result Improved convergence rates in various distances for neural networks.
Gradient descent memorizes many Gaussians efficiently.
problem Memorizing many Gaussians with minimal parameters.
method Gradient descent on a depth-two neural network.
result One step of gradient descent memorizes $Ω\left(\frac{dq}{\log^4(d)}
ight)$ Gaussians.