arXiv research

Deep belief networks can approximate any multivariate density with binary hidden units.

Paper proposes a novel spectral approach to learn binary latent variable models.

Improves causal graph learning on dependent binary data.

Feed-forward nets fail to learn equality relations, but adding DR units helps.

We establish upper bounds for the minimal number of hidden units for which a binary stochastic feedforward network with sigmoid activation probabilities and a single hidden layer is a universal approximator of Markov kernels. We show that each possible probabilistic assignment of the states of $n$ output units, given t…

Reduced parameter space improves neural network training efficiency.

The mesoscopic organization of complex systems, from financial markets to the brain, is an intermediate between the microscopic dynamics of individual units (stocks or neurons, in the mentioned cases), and the macroscopic dynamics of the system as a whole. The organization is determined by "communities" of units whose …

Binary testing for softmax models requires many samples, similar to leverage score models.

Model neural plasticity as binary optimization to dynamically activate or deactivate network units.

This study uses neural networks to solve interpolation problems with sparse, infinitely wide layers.

A note on extending Chernoff bound for unit interval random variables.

Combining interpretability and stability methods improves DNN robustness.

A neural network with a single hidden layer can't represent certain multivariable functions.

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,…

A new method sparsifies neural networks using stochastic binary optimization.

The paper provides theoretical guarantees for neural network-based anomaly detection.

A new scoring function improves ordinal regression by combining binary predictions.

Dual neural network architecture improves accuracy and interpretability.

Ward2ICU dataset protects patient privacy while generating synthetic ICU transitions data.

Neural network combines GRU and SVM for better intrusion detection.

New method improves accuracy of quantized neural networks.

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…

Lossy compression of statistical data using quantum annealing.

The paper uses permutation representations to visualize group extensions and subgroups.

Machine learning predicts movie genres from summaries with high accuracy.

We improve recently published results about resources of Restricted Boltzmann Machines (RBM) and Deep Belief Networks (DBN) required to make them Universal Approximators. We show that any distribution p on the set of binary vectors of length n can be arbitrarily well approximated by an RBM with k-1 hidden units, where …

The beta-Bernoulli process provides a Bayesian nonparametric prior for models involving collections of binary-valued features. A draw from the beta process yields an infinite collection of probabilities in the unit interval, and a draw from the Bernoulli process turns these into binary-valued features. Recent work has …

PoET-BiN reduces power consumption in neural networks on embedded devices.

Paper proposes binary-valued gates for better LSTM training.

Analyzes how ReLU affects neural network capacity and solution space geometry.

SAPSAM trains CNNs on lung CTs with binary labels, improving CPA detection and localization.

Stochastic binary hidden units in a multi-layer perceptron (MLP) network give at least three potential benefits when compared to deterministic MLP networks. (1) They allow to learn one-to-many type of mappings. (2) They can be used in structured prediction problems, where modeling the internal structure of the output i…

Background: Statistical mechanics results (Dauphin et al. (2014); Choromanska et al. (2015)) suggest that local minima with high error are exponentially rare in high dimensions. However, to prove low error guarantees for Multilayer Neural Networks (MNNs), previous works so far required either a heavily modified MNN mod…

GRU-D detects age-specific missing patterns in vital signs.

2-bit quantization improves RNN performance on resource-limited devices.

UniPhyNet improves cognitive load classification accuracy using EEG, ECG, and EDA signals.

CRBM extracts speech features from complex spectra directly.

New approach estimates personalized treatment effects using surrogate losses.

M3E2 neural network estimates multiple treatment effects.

Can we make Bayesian posterior MCMC sampling more efficient when faced with very large datasets? We argue that computing the likelihood for N datapoints in the Metropolis-Hastings (MH) test to reach a single binary decision is computationally inefficient. We introduce an approximate MH rule based on a sequential hypoth…

Framework quantifies neural network robustness to parameter quantization.

Study causal inference under specific sampling methods with monotonicity assumptions.

Thompson Sampling achieves independent regret bounds for logistic bandits.

This paper resolves BIHT convergence, showing normalization is not necessary in noiseless settings but crucial for robustness.

DNPUs improve neural network performance with high-capacity nanoelectronic nodes.

Improves classifier evaluation by aligning with Total Classification Cost.

Methods from convex optimization are widely used as building blocks for deep learning algorithms. However, the reasons for their empirical success are unclear, since modern convolutional networks (convnets), incorporating rectifier units and max-pooling, are neither smooth nor convex. Standard guarantees therefore do n…

Paper tackles learning ReLU networks for binary classification with linearly separable data.