◐ Latest Activity

Image and Video Processing

A web-based GAN for dimensionality reduction

Interactive Latent Interpolation on MNIST Dataset

In this paper, we discuss the potential of dimensionality reduction with a web-based use of generative adversarial networks (gan).

This is primarily useful when coupled with dimensionality reduction as an effective application of our algorithm.

Cold Case: The Lost MNIST Digits

Although the popular MNIST dataset [LeCun et al., 1994] is derived from the NIST database [Grother and Hanaoka, 1995], the precise processing steps for this derivation have been lost to time. We propo

More...

Typography-MNIST (TMNIST)

Typography-MNIST (TMNIST): an MNIST-Style Image Dataset to Categorize Glyphs and Font-Styles

We present a dataset comprising of 565,292mnist-style grayscale images representing 1,812 unique glyphs in varied styles of 1,355 google-fonts.

The glyph-list contains common characters from over 150of the modern and historical language scripts with symbol sets, and each font-style represents varying subsets of the total unique glyphs.

CC BY-SA

A Benchmark for Multi-Class Object Recognition with Deep Learning

Overhead MNIST: A Benchmark Satellite Dataset

This research presents an overhead view of 10 important objects and follows the general formatting requirements of the most popular machine learning task: digit recognition with the mnist.

This dataset offers a public benchmark extractedfrom over a million human-labelled and curated examples.

The Overhead-MNIST Image Classification Benchmark

Artificial Intelligence

Overhead-MNIST: Machine Learning Baselines for Image Classification

The overhead-mnist dataset is a collection of satellite images similar in styleto the ubiquitous mnist hand-written digits found in the machine learningliterature.

Twenty-three machine learning algorithms were trained then scored to establish baseline comparison metrics and to select an image classificationalgorithm worthy of embedding into mission-critical satellite imaging systems.

EMNIST: an extension of MNIST to handwritten letters

The MNIST dataset has become a standard benchmark for learning, classification and computer vision systems. Contributing to its widespread adoption are the understandable and intuitive nature of the t

More...

Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1

We introduce a method to train Binarized Neural Networks (BNNs) - neural networks with binary weights and activations at run-time. At training-time the binary weights and activations are used for comp

More...

Neurons and Cognition

Can Single Neurons Solve MNIST? The Computational Power of Biological Dendritic Trees

Physiological experiments have highlighted how the dendrites of biological neurons can nonlinearly process distributed synaptic inputs. This is in stark contrast to units in artificial neural networks

More...

LightLayers: Reducing the Complexity of Deep Neural Networks

LightLayers: Parameter Efficient Dense and Convolutional Layers for Image Classification

We propose a method for reducing the number of trainable parameters in deep neural networks (dnn).

The proposed method is based on matrix factorization to reduce the complexity of the dnn models resulting into lightweight dnnmodels that require less computational power, without much loss in the accuracy.

Oracle-MNIST: a Realistic Image Dataset for Benchmarking Machine Learning Algorithms

The Oracle-MNIST Dataset

We introduce the oracle-mnist dataset, comprising of grayscale images of 30,222 ancient characters from 10 categories, for benchmarking pattern classification, with particular challenges on image noise and distortion.

The training set totally consists of 27,222 images, and the testset contains 300 images per class.

Adversarial Attacks on Variational Autoencoders

Defending Variational Autoencoders from Adversarial Attacks with MCMC

Variational autoencoders (vaes) are deep generative models used in various domains.

However, one can easily fool vaes to produce unexpected latent representations and reconstructions for a visually slightly modified input.

Quantum Physics

Single-Photon Image Classification

Quantum computing-based machine learning mainly focuses on quantum computing hardware that is experimentally challenging to realize due to requiring quantum gates that operate at very low temperature.

More...

Renormalized Sparse Neural Network Pruning

Large neural networks are heavily over-parameterized. This is done because it improves training to optimality. However once the network is trained, this means many parameters can be zeroed, or pruned,

More...

Towards the first adversarially robust neural network model on MNIST

Despite much effort, deep neural networks remain highly susceptible to tiny input perturbations and even for MNIST, one of the most common toy datasets in computer vision, no neural network model exis

More...

Predicting with High Correlation Features

It has been shown that instead of learning actual object features, deep networks tend to exploit non-robust (spurious) discriminative features that are shared between training and test sets. Therefore

More...

High Energy Physics - Experiment

High Energy Physics - Phenomenology

Data Analysis

Computational Physics

Generating Sparse Representations of MNIST Digits and Particles

Graph Generative Adversarial Networks for Sparse Data Generation in High Energy Physics

We develop a graph generative adversarial network to generate sparse representations of data sets like those produced at the cern large hadron collider (lhc).

We demonstrate this approach by training on and generating sparse representations of handwritten digit images and jets of particles in proton-proton collisions like those at the lhc.