The study examines deep convolutional neural networks and their learning ability.
problem Understanding the learning ability of deep convolutional neural networks (DCNNs).
method Examines DCNNs under both underparameterized and overparameterized settings, using a novel network deepening scheme.
result Establishes the first learning rates of underparameterized DCNNs and shows how adding layers can create interpolating DCNNs with good learning rates.
Two new criteria help understand the advantage of deep neural networks.
problem Understanding the advantage of deepening neural networks.
method Proposed two new criteria to evaluate the expressivity of functions computable by deep neural networks.
result Increasing layers is more effective than increasing units in improving the expressivity of deep neural networks.
Ricci flow deforms the Riemannian metric proportionally to the curvature, such that the curvature evolves according to a heat diffusion process and eventually becomes constant everywhere. Ricci flow has demonstrated its great potential by solving various problems in many fields, which can be hardly handled by alternati…
Study of autocorrelation times in neural MCMC simulations for the 2D Ising model.
problem Estimating autocorrelation times in Neural Markov Chain Monte Carlo simulations.
method Analytical and empirical methods to estimate autocorrelation times, proposing new loss functions and training schemes.
result Proposed new loss functions and training schemes that improve autocorrelation times in neural MCMC simulations.
Deep neural networks and in particular, deep neural classifiers have become an integral part of many modern applications. Despite their practical success, we still have limited knowledge of how they work and the demand for such an understanding is evergrowing. In this regard, one crucial aspect of deep neural network c…
Regularized policies are robust to adversarial rewards.
problem Understanding the effects of regularization on policy exploration and robustness.
method Using Fenchel duality to derive the dual problem of the regularized RL objective, showing the optimal policy is robust to adversarial rewards.
result Regularized policies are optimal for a reinforcement learning problem under adversarial reward conditions.
Paper introduces vector-valued variation spaces for multi-output neural networks.
problem Understanding and optimizing multi-output neural networks.
method Development of vector-valued variation spaces and representer theorem.
result Novel bounds for layer widths in deep networks and a convex optimization method for compression.
Unified framework DDNs for multi-label classification, improving inference efficiency.
problem Efficient inference for multi-label classification with dependency networks.
method Combining dependency networks and deep learning, proposing novel inference schemes.
result Novel inference schemes outperform basic neural architectures and Markov networks.
Paper introduces Cycles Protocol to integrate trade credit into market clearing.
problem Liquidity embedded in trade credit outside formal settlement infrastructures.
method Distributed, multilateral clearing mechanism based on double-entry accounting.
result Cycles Protocol maximizes balance sheet compression without redistributing counterparty risk.
Performance of neural networks can be significantly improved by encoding known invariance for particular tasks. Many image classification tasks, such as those related to cellular imaging, exhibit invariance to rotation. We present a novel scheme using the magnitude response of the 2D-discrete-Fourier transform (2D-DFT)…
Derives equations for capital deepening in a competitive economy without assuming a production function.
problem Understanding capital deepening and firm survival in a competitive economy.
method Derives equations of motion from accounting identities, without assuming a production function. Uses four coupled relaxation equations to govern capital productivity, labor share, and new investment productivity.
result A 1% improvement in new-capital productivity nearly doubles the aggregate growth rate within one capital lifetime.
Proposes a new CG interpretation of neural networks for better theoretical analysis.
problem Lack of theoretical analysis in neural networks interpretation.
method Interprets neural networks as chain graphs and feed-forward as approximate inference.
result Provides novel theoretical support and insights for various neural network techniques.
Diagonal linear networks converge to lasso regularization path during training.
problem Understanding the regularization behavior of diagonal linear networks.
method Analyzing the training trajectory of diagonal linear networks and comparing it to the lasso regularization path.
result The training trajectory of diagonal linear networks is closely related to the lasso regularization path.
Develops statistical guarantees for neural networks with regularization.
problem Lack of comprehensive mathematical theories for neural networks.
method General statistical guarantee for least-squares with regularizers.
result Prediction error increases sub-linearly in layers, logarithmically in parameters.
Graph structured data are abundant in the real world. Among different graph types, directed acyclic graphs (DAGs) are of particular interest to machine learning researchers, as many machine learning models are realized as computations on DAGs, including neural networks and Bayesian networks. In this paper, we study dee…
Proposes dynamic channel pruning during neural network training.
problem Pruning neural networks during training to reduce computational cost and improve efficiency.
method Dynamic channel propagation to update channel utility values and selectively prune channels.
result Our scheme trains and prunes neural networks simultaneously, achieving superior performance.
The properties of q-dependent cross-correlation matrices of stock market have been analyzed by using the random matrix theory and complex network. The correlation structures of the fluctuations at different magnitudes have unique properties. The cross-correlations among small fluctuations are much stronger than those a…
FedOS tackles challenges in federated learning by using open-set learning.
problem Challenges in federated learning due to data locality and privacy constraints.
method Introduces open-set learning to stabilize training in federated learning.
result Demonstrates improved model performance through open-set learning.
The paper explores how to find relevant vertices in one graph using another graph's attributes and structure.
problem Finding relevant vertices in one graph using another graph's attributes and structure.
method Theoretical and practical exploration of vertex nomination schemes that leverage both content (edge and vertex attributes) and context (network topology).
result Necessary and sufficient conditions for schemes that use both content and context to outperform those using only one.
MPNNs generalize poorly, study reviews current research.
problem Exploring generalization abilities of MPNNs.
method Systematic review of existing literature.
result Limited understanding of MPNNs' generalization abilities.
Bayesian network learns data invariances without augmentation.
problem Learning invariances in neural networks without manual design.
method Bayesian approach infers weight-sharing schemes from data.
result Model outperforms non-invariant networks on specific tasks.
A new privacy-preserving deep learning scheme for asymmetrically collaborative machine learning.
problem Privacy and efficiency in collaborative machine learning across different data owners.
method Decomposes neural network steps for privacy-preserving training; novel protocol for information leakage.
result Efficient training with stable performance and significant speedup.
Long Short-Term Memory (LSTM) is a popular approach to boosting the ability of Recurrent Neural Networks to store longer term temporal information. The capacity of an LSTM network can be increased by widening and adding layers. However, usually the former introduces additional parameters, while the latter increases the…
Novel oracle-type inequality for logistic loss in DNNs achieves sharp convergence rates.
problem Generalization analysis for binary classification with DNNs and logistic loss.
method Established an oracle-type inequality to handle the boundedness of the target function.
result Optimal convergence rates for fully connected ReLU DNN classifiers trained with logistic loss.
Message-passing neural networks (MPNNs) have been successfully applied to representation learning on graphs in a variety of real-world applications. However, two fundamental weaknesses of MPNNs' aggregators limit their ability to represent graph-structured data: losing the structural information of nodes in neighborhoo…
Bayesian networks are probabilistic graphical models widely employed to understand dependencies in high dimensional data, and even to facilitate causal discovery. Learning the underlying network structure, which is encoded as a directed acyclic graph (DAG) is highly challenging mainly due to the vast number of possible…
Paper proposes new method for time series confidence intervals using LSTM.
problem Constructing accurate confidence intervals for multivariate time series.
method Uses Long Short Term Memory Network (LSTM) and novel block bootstrap techniques.
result Demonstrates improved accuracy in constructing confidence intervals.
Paper introduces a new IS scheme for estimating distribution tails of complex models.
problem Scalability and feasibility issues in traditional IS schemes for rich models.
method Develops a self-structuring IS approach guided by large deviations principles.
result First to achieve asymptotically optimal variance reduction across various multivariate distributions.
Novel framework for policy optimization with general parameterization and linear convergence.
problem Lack of theoretical guarantees for policy optimization with general parameterization schemes.
method Mirror descent approach for policy optimization with general parameterization.
result First result of linear convergence for policy-gradient-based method with general parameterization.
Deep Neural Networks (DNNs) are usually over-parameterized, causing excessive memory and interconnection cost on the hardware platform. Existing pruning approaches remove secondary parameters at the end of training to reduce the model size; but without exploiting the intrinsic network property, they still require the f…
Unified analysis of DLNs using DMFT reveals dynamics of loss convergence and generalization trade-offs.
problem Understanding the overall dynamics of diagonal linear networks (DLNs) in neural network training.
method Dynamical Mean-Field Theory (DMFT) applied to DLNs.
result Derives low-dimensional effective process capturing high-dimensional gradient flow dynamics.
Deep learning accelerates Monte Carlo SDE simulations with large time steps.
problem Accurate simulation of SDEs with large time steps.
method Polynomial chaos expansion with neural network learned stochastic collocation points.
result Data-driven scheme achieves strong convergence in Monte Carlo simulations.
To deepen our understanding of graph neural networks, we investigate the representation power of Graph Convolutional Networks (GCN) through the looking glass of graph moments, a key property of graph topology encoding path of various lengths. We find that GCNs are rather restrictive in learning graph moments. Without c…
Recurrent Neural Networks (RNNs) are designed to handle sequential data but suffer from vanishing or exploding gradients. Recent work on Unitary Recurrent Neural Networks (uRNNs) have been used to address this issue and in some cases, exceed the capabilities of Long Short-Term Memory networks (LSTMs). We propose a simp…
Optimal resource allocation is a fundamental challenge for dense and heterogeneous wireless networks with massive wireless connections. Because of the non-convex nature of the optimization problem, it is computationally demanding to obtain the optimal resource allocation. Recently, deep reinforcement learning (DRL) has…
We propose a distributed approach to train deep neural networks (DNNs), which has guaranteed convergence theoretically and great scalability empirically: close to 6 times faster on instance of ImageNet data set when run with 6 machines. The proposed scheme is close to optimally scalable in terms of number of machines, …
We seek to deepen understanding of the micro-foundations of institutionalization while contributing to a sociological theory of markets by investigating the puzzle of price bubbles in financial markets. We find that such markets, despite textbook conditions of high efficiency -- perfect information, atomistic agents, n…
Develops a numerical scheme for solving path-dependent FBSDEs and PDEs.
problem Solving path-dependent FBSDEs and PDEs numerically.
method Picard iteration method for FBSDEs, concentration inequality for estimator, supervised learning with neural networks for PDEs.
result Proves convergence and rate of convergence for the Picard iteration method.
The incremental aggregated gradient algorithm is popular in network optimization and machine learning research. However, the current convergence results require the objective function to be strongly convex. And the existing convergence rates are also limited to linear convergence. Due to the mathematical techniques, th…
Representing maritime traffic patterns and detecting anomalies from them are key to vessel monitoring and maritime situational awareness. We propose a novel approach -- referred to as GeoTrackNet -- for maritime anomaly detection from AIS data streams. Our model exploits state-of-the-art neural network schemes to learn…
AWARE improves graph prediction by aggregating walks with attention schemes.
problem Improving graph prediction accuracy using walk aggregation.
method Integrates attention schemes into walk-aggregating GNNs.
result AWARE outperforms existing methods in graph-level prediction tasks.
A new autoencoder architecture captures multiscale data.
problem Multiscale spatio-temporal data representation.
method Integrates multigrid methods, convolutional autoencoders, and transfer learning.
result Adaptive, hierarchical architecture captures different scaled features dynamically.
A new neural network initialization method is proposed for faster and more accurate training.
problem Efficient initialization for training multi-layer feedforward neural networks.
method Initialization based on Stein's identity, using eigenvectors of cross-moment matrix.
result The SteinGLM method is faster and more accurate than other initialization methods.
Variational auto-encoders (VAEs) are a popular and powerful deep generative model. Previous works on VAEs have assumed a factorized likelihood model, whereby the output uncertainty of each pixel is assumed to be independent. This approximation is clearly limited as demonstrated by observing a residual image from a VAE …
Novel IMEX scheme solves financial PDEs with mixed derivatives.
problem Numerical approximations for financial PDEs with mixed derivatives.
method Second order finite volume IMEX Runge-Kutta scheme.
result Achieves true second order convergence with non-regular initial conditions.
New insights into correntropy-based regression reveal robustness and unified approaches.
problem Learning robust regression functions under additive noise.
method Minimum distance estimation and conditional mean, mode, median functions.
result Unified approach to conditional mean, mode, and median functions.
Study on SGD for overparameterized neural networks, focusing on convergence rates.
problem Understanding convergence rates of SGD in overparameterized two-layer neural networks.
method Combines NTK approximation with RKHS analysis to explore SGD dynamics.
result Established sharp convergence rates for SGD in overparameterized two-layer neural networks.
Paper addresses FL over MAC with DP constraints, proposing a novel consensus scheme.
problem Federated learning over a multiple access channel with differential privacy constraints.
method Proposes a novel consensus scheme using digital distributed stochastic gradient descent (D-DSGD) with artificial noise to preserve DP.
result Demonstrates improved convergence rate and DP level for a given MAC capacity.