
Semisupervised learning using convolutional neural networks for honey bee colony classification
Semisupervised learning using convolutional neural networks for honey bee colony classification – The recent advances in deep learning allow deep learning techniques to capture rich information, but are often only seen in the context of a single single image, which is difficult to interpret in largescale systems. In this work we present a fully automatic […]

Recurrent Neural Networks for Visual Recognition
Recurrent Neural Networks for Visual Recognition – We present a new approach for the automatic detection of human hand gestures. The proposed approach is based on neural network models trained on a sequence of hand gestures. A recurrent neural network (RNN) is trained to predict the presence of a hand as the human visualizes it […]

Generative Deep Episodic Modeling
Generative Deep Episodic Modeling – Deep neural networks, or more broadly, learning models with deep embeddings, enable a wide range of applications on a variety of levels: from biomedical data to language modeling. In this work, we study the feasibility and performance of learning models on structured data and on unstructured language models, and compare […]

Unsupervised Semantic Segmentation of Long VideoCSR
Unsupervised Semantic Segmentation of Long VideoCSR – This paper presents deep learning with semisupervised semantic segmentation (SRD) using a fully contextualized deep CNN for learning the semantic segmentation task from videos of subjects. Our SRD network takes the input video as input and transforms it into semantic segmentation vector (i.e. training a deep neural network). […]

Boosting the Effects of Multiple Inputs
Boosting the Effects of Multiple Inputs – We revisit the classical problem of the maximum likelihood estimation problem, which is one of the most wellstudied problems in physics. Since the problem is NPhard, the best heuristic algorithms for solving it (i.e., the optimal solution) need an optimization strategy that avoids the sparsmodal optimization problem. This […]

Variational Approximation via Approximations of Approximate Inference
Variational Approximation via Approximations of Approximate Inference – This paper proposes a novel method for extracting useful information from noisy data by fitting a posterior distribution to the expected expected time to information transmission in terms of the time it takes to respond to a given data frame on a given data set. By combining […]

DeepKSPD: Learning to detect unusual motion patterns in videos
DeepKSPD: Learning to detect unusual motion patterns in videos – In a video game, the players have chosen various actions which could be useful in achieving their goals or in avoiding mistakes. Using a realworld video simulation as an example, we show that the player’s actions in games are not relevant for a certain goal: […]

A Novel Unsupervised Learning Approach for Multiple Attractor Learning on Graphs
A Novel Unsupervised Learning Approach for Multiple Attractor Learning on Graphs – In this paper, for the first time, we propose a general algorithm for multisource labeling systems and label extraction through a network of nodes that simultaneously learns features from a few classes and then aggregates them. Such learning is very challenging with few […]

On the Generalization of Randomized Loss Functions in Deep Learning
On the Generalization of Randomized Loss Functions in Deep Learning – The problem of learning how to classify a collection of images from a movie can be viewed as a continuous learning problem from a supervised learning setting. Unfortunately, the training objective does not have a principled way of predicting whether the image (or the […]

On Detecting Similar Languages in Text in Hindi
On Detecting Similar Languages in Text in Hindi – The purpose of this paper is to develop a framework that enables the automated verification and classification of two commonly recognised linguistic terms in Hindi text, i.e., isatiya and indian. In this paper, we use the phrase ‘indian lang’ to categorise isatiyaneighbourhood as Hindi in terms […]