Study of infinitely deep but narrow neural networks using NTK theory.
problem Analyzing the role of depth in deep learning with overparameterized networks.
method Infinite-depth limit analysis of MLP and CNN using Neural Tangent Kernel (NTK) theory.
result Established trainability guarantee for infinitely deep but narrow neural networks.
We show that deep narrow Boltzmann machines are universal approximators of probability distributions on the activities of their visible units, provided they have sufficiently many hidden layers, each containing the same number of units as the visible layer. We show that, within certain parameter domains, deep Boltzmann…
Deep narrow networks can approximate any continuous function.
problem Approximating continuous functions with neural networks of bounded width and arbitrary depth.
method Showed neural networks of arbitrary depth, width n+m+2, and activation function ρ is dense in C(K;Rm) for K⊆Rn with K compact. result Neural networks of bounded width and arbitrary depth can approximate any continuous function.
Recent theoretical work has demonstrated that deep neural networks have superior performance over shallow networks, but their training is more difficult, e.g., they suffer from the vanishing gradient problem. This problem can be typically resolved by the rectified linear unit (ReLU) activation. However, here we show th…
Study proves deep narrow RNNs can approximate any function, with minimum width independent of data length.
problem Proving universality of deep narrow RNNs with bounded widths.
method Analyzing RNNs as dynamical systems, proving universality for deep narrow structures with specific widths.
result Minimum width for universality of deep narrow RNNs is independent of data length.
We show that for neural network functions that have width less or equal to the input dimension all connected components of decision regions are unbounded. The result holds for continuous and strictly monotonic activation functions as well as for the ReLU activation function. This complements recent results on approxima…
Improved bounds on neural network expressivity.
problem Understanding neural network expressivity and approximation capabilities.
method Improved bounds on the maximal number of linear regions of ReLU-networks.
result New insights into the expressivity of neural networks.
Embedding principle explains loss landscape of deep neural networks.
problem Understanding the structure of loss landscapes in deep neural networks.
method Proposed an embedding principle that critical points of narrower DNNs can be embedded to critical points of wider DNNs.
result Wide DNNs are often attracted by highly-degenerate critical points embedded from narrower DNNs.
Study compares random and learned features in deep Bayesian linear models.
problem Understanding how feature learning affects generalization in deep learning.
method Comparing deep random feature models to deep networks with trained layers.
result Random feature models can display double-descent behavior, while deep networks do not.
New approach finds minimum width for deep, narrow MLPs.
problem Finding the minimum width for deep, narrow MLPs to approximate continuous functions.
method Proposes a framework to simplify finding minimum width into determining a geometrical function w(dx,dy) based on input and output dimensions. result Proves that w(dx,dy) equals the optimal minimum width for deep, narrow MLPs to achieve universality. We review recent results about the maximal values of the Kullback-Leibler information divergence from statistical models defined by neural networks, including naive Bayes models, restricted Boltzmann machines, deep belief networks, and various classes of exponential families. We illustrate approaches to compute the max…
We generalize recent theoretical work on the minimal number of layers of narrow deep belief networks that can approximate any probability distribution on the states of their visible units arbitrarily well. We relax the setting of binary units (Sutskever and Hinton, 2008; Le Roux and Bengio, 2008, 2010; Montúfar and Ay,…
Deep learning uncovers patterns between knot types.
problem Discovering connections between combinatorial and hyperbolic knot invariants.
method Statistical approach using linear regression and deep learning.
result Revealed empirical connections between knot types.
Wide neural networks with narrow bottlenecks behave like deep Gaussian processes.
problem Understanding the behavior of neural networks with narrow layers in the wide limit.
method Analyzing the wide limit of BNNs with narrow bottlenecks, showing they behave like a composition of GPs.
result Wide neural networks with narrow bottlenecks form a composition of GPs, termed a bottleneck NNGP.
This paper improves deep neural network approximation for fully connected networks, achieving optimal convergence rates.
problem Improving approximation of fully connected deep neural networks for optimal convergence rates.
method Deriving approximation bounds specifically for a narrower fully connected deep neural network.
result Achieves an optimal rate (up to a logarithmic factor) for fully connected deep neural networks.
Statistical physics explains deep learning's feature learning capacity.
problem Understanding neural networks' ability to learn complex features.
method Study of a multi-layer perceptron in the interpolation regime.
result Optimal learning requires specialization across layers and neurons.
We prove that for an L-layer fully-connected linear neural network, if the width of every hidden layer is Ω~(L⋅r⋅dout⋅κ3), where r and κ are the rank and the condition number of the input data, and dout is the output dimension, then gradient descent with Gaussi…
Complex-valued neural networks can approximate any continuous function with bounded widths and depths.
problem Approximating continuous functions with complex-valued neural networks of bounded widths and depths.
method Analyzing activation functions and proving universality for complex-valued networks.
result Deep narrow complex-valued networks are universal if and only if their activation function is neither holomorphic, nor antiholomorphic, nor R-affine. We demonstrate that a very deep ResNet with stacked modules with one neuron per hidden layer and ReLU activation functions can uniformly approximate any Lebesgue integrable function in d dimensions, i.e. ℓ1(Rd). Because of the identity mapping inherent to ResNets, our network has alternating layers of…
Neural networks can approximate functions uniformly across various measures.
problem Universal approximation of functions across different probability measures.
method Proving neural networks are dense in Orlicz spaces, extending classical theorems.
result Neural networks uniformly approximate functions for weakly compact families of measures.
Wide networks are often believed to have a nice optimization landscape, but what rigorous results can we prove? To understand the benefit of width, it is important to identify the difference between wide and narrow networks. In this work, we prove that from narrow to wide networks, there is a phase transition from havi…
Tilting loss functions improves machine learning performance.
problem Improving machine learning models, especially in under- and over-parameterized networks.
method Using evolving loss functions that emphasize different classes cyclically.
result Dynamical loss functions lead to better generalization and stability in training.
Proposes efficient training method for deep thin networks.
problem Deploying deep learning models with accuracy and compactness.
method Three-stage method: widen, warm up, fine tune.
result Deep thin networks trained with method outperform standard deep networks.
Compression affects deep networks differently, impacting underrepresented data points.
problem Disparate impact of compression on different classes and images.
method Analysis of deep neural network pruning and quantization effects.
result Compression disproportionately impacts model performance on underrepresented data points.
This paper presents a method to automatically generate high-quality prediction intervals for neural networks.
problem Accurate uncertainty quantification for deep learning models in real-world applications.
method Dual neural network approach with a novel loss function to balance prediction interval width and coverage.
result Our method produces significantly narrower prediction intervals with higher probability coverage compared to state-of-the-art methods.
Study shows DNNs can recover functions with fewer samples than model parameters at overparameterization.
problem Determining reliable function recovery in overparameterized deep neural networks.
method Introducing 'local linear recovery' (LLR) and proving upper bounds on sample sizes for recovery.
result Upper bounds on optimistic sample sizes for function recovery in overparameterized DNNs are achieved.
New model explains deep learning performance at large learning rates.
problem Understanding deep learning performance at different learning rates.
method Developed neural networks with solvable training dynamics.
result Large learning rates lead to convergence to flatter minima.
New algorithm ensures global convergence in deep neural networks beyond NTK regime.
problem Existing global convergence guarantees do not apply to practical deep networks.
method Proposes an algorithm with global convergence guarantees under the expressivity condition.
result Algorithm ensures global convergence in practical settings beyond NTK regime.
Deep learning architectures have proved versatile in a number of drug discovery applications, including the modelling of in vitro compound activity. While controlling for prediction confidence is essential to increase the trust, interpretability and usefulness of virtual screening models in drug discovery, techniques t…
HLOB predicts mid-price changes in L.O.Bs using deep learning.
problem Forecasting mid-price changes in Limit Order Books.
method HLOB uses a deep learning model with an Information Filtering Network and Homological Convolutional Neural Networks.
result HLOB outperforms state-of-the-art models in real-world datasets.
We can compare the expressiveness of neural networks that use rectified linear units (ReLUs) by the number of linear regions, which reflect the number of pieces of the piecewise linear functions modeled by such networks. However, enumerating these regions is prohibitive and the known analytical bounds are identical for…
Wide neural networks can degrade performance, contrary to conventional wisdom.
problem Understanding the limitations of increasing network width in neural networks.
method Using Deep Gaussian Processes to decouple capacity and width, analyzing their effects on representational power and non-Gaussianity.
result Wide neural networks can become less adaptable and more Gaussian, leading to performance degradation.
New framework analyzes deep learning optimization with finite width networks, revealing generalization gaps and excess risks.
problem Analyzing generalization error of deep learning with finite width networks.
method Formulating neural network training as transportation map estimation and analyzing via infinite dimensional Langevin dynamics.
result Achieves fast learning rate and minimax optimal rates for classification and regression problems.
In (\cite{zhang2014nonlinear,zhang2014nonlinear2}), we have viewed machine learning as a coding and dimensionality reduction problem, and further proposed a simple unsupervised dimensionality reduction method, entitled deep distributed random samplings (DDRS). In this paper, we further extend it to supervised learning …
Large learning rates improve generalization, but optimal ranges are narrower than previously thought.
problem Optimizing learning rates for neural network training.
method Detailed exploration of learning rate ranges in a simplified setup, validating findings in a practical setting.
result Optimal learning rate ranges are significantly narrower than previously assumed.
Deep neural networks can classify any dataset with exact solutions.
problem Classifying any dataset with multilayer perceptrons.
method Constructive analysis of multilayer perceptrons with exact solutions.
result Explicit constructions of MLPs for classification and function approximation.
This paper considers the generation of prediction intervals (PIs) by neural networks for quantifying uncertainty in regression tasks. It is axiomatic that high-quality PIs should be as narrow as possible, whilst capturing a specified portion of data. We derive a loss function directly from this axiom that requires no d…
With ever-increasing computational demand for deep learning, it is critical to investigate the implications of the numeric representation and precision of DNN model weights and activations on computational efficiency. In this work, we explore unconventional narrow-precision floating-point representations as it relates …
As deep neural networks continue to revolutionize various application domains, there is increasing interest in making these powerful models more understandable and interpretable, and narrowing down the causes of good and bad predictions. We focus on recurrent neural networks (RNNs), state of the art models in speech re…
Adversarial attacks reduce deep learning beam selection performance in mmWave 5G.
problem Adversarial attacks degrade deep learning-based beam selection in mmWave 5G.
method Generates adversarial perturbations to RSS inputs to manipulate DNN predictions.
result Significant reduction in IA performance due to adversarial perturbations.
Finding minima of a real valued non-convex function over a high dimensional space is a major challenge in science. We provide evidence that some such functions that are defined on high dimensional domains have a narrow band of values whose pre-image contains the bulk of its critical points. This is in contrast with the…
In this paper, we propose a simple but effective method for training neural networks with a limited amount of training data. Our approach inherits the idea of knowledge distillation that transfers knowledge from a deep or wide reference model to a shallow or narrow target model. The proposed method employs this idea to…
Unified approach to continual learning using generative replay and open set recognition.
problem Catastrophic interference and recognition of out-of-distribution data in deep neural networks.
method Probabilistic approach based on variational inference in a deep autoencoder model, using generative replay and open set recognition.
result The approach significantly alleviates catastrophic interference and distinguishes out-of-distribution data.
Deep hedging uses RL to minimize risk in financial markets.
problem Minimizing risk in financial markets using reinforcement learning.
method Trains a neural network policy via Monte Carlo simulation and stochastic gradient descent.
result Deep hedging algorithm falls within the RL category.
Bayesian Quadrature improves ensembling for neural networks with dispersed likelihood peaks.
problem Ensembling neural networks struggles with dispersed, narrow peaks in likelihood surfaces.
method Uses Bayesian Quadrature to construct weighted ensembles of architectures.
result Empirically outperforms state-of-the-art baselines in test likelihood, accuracy, and expected calibration error.
DOFEN improves DNN performance on tabular data benchmarks.
problem DOFEN tackles the performance gap between DNNs and tree-based models on tabular data.
method DOFEN uses a two-level rODT forest ensembling process inspired by oblivious decision trees.
result DOFEN achieves state-of-the-art results on the Tabular Benchmark.
A new method reduces memory requirements for Graph Transformers by sparsely training a network.
problem Quadratic memory complexity in Graph Transformers limits their scalability to large graphs.
method Spexphormer: trains a narrow network on augmented graph, then uses only active connections in a wider network.
result Spexphormer achieves good performance with drastically reduced memory requirements.
New findings show depth is more important than width in neural networks.
problem Understanding the role of width and depth in neural networks.
method Constructed networks with bounded weights and width at most d+2, showing depth plays a more significant role.
result Depth is more important than width in the expressive power of neural networks.