In this story, I am going to talk about one of the trending topics in different domains of machine learning and deep learning, from natural language processing to computer vision: Zero and Few Shot Learning Few-shot Learning As we know, in supervised classifiers, different models tries to learn features in the data…

DistilBERT is a language model designed by HuggingFace to achieve BERT model performance with fewer parameters, resulting in a quicker model. Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf demonstrated that a BERT model may be reduced in size by 40% while preserving 97% of its language processing skills…

From computer vision to speech recognition to Natural Language Processing (NLP) and beyond, neural networks excel in a wide range of applications. …

Word embedding is a phrase used in natural language processing (NLP) to describe the representation of words for text analysis, which is often in the form of a real-valued vector that encodes the meaning of the word and predicts the meaning of words that are close in the vector space…

In this article, I will introduce you to various ways of text preprocessing which is one the most important levels of and NLP project. …

Object detection is unquestionably one of the most significant challenges in the field of computer vision. One method for this job is R-CNN, which was introduced by Ross Girshick, Jeff Donahue, Trevor Darrell, and Jitendra Malik in “Rich feature hierarchies for accurate object recognition and semantic segmentation”. …

Authors: William Chan, Navdeep Jaitly, Quoc V. Le, Oriol Vinyals Published On: 2015 Before starting the review there are some concepts which is good to learn them first. Pyramidal recurrent unit This is based on Pyramidal Recurrent Unit for Language Modeling paper by Sachin Metha, Rik Koncel-Kedziorski, Mohammad Rastegari and Hannaneh Hajishirzi. PRUs…

A recurrent neural network (RNN) is a class of artificial neural networks where connections between nodes form a directed graph along a temporal sequence. This allows it to exhibit temporal dynamic behavior. …

Author : Steffen Schneider, Alexi Baevski, Ronan Collobertm Micheal Auli Overview Unsupervised pre-training for representation learning for speech recognition We pre-train a simple multi-layer convolutional neural network optimized via a noise contrastive binary classification task Experiments on WSJ reduce WER of strong character-based log-mel filter bank baseline by up to 46%…