Deep Learning Overview

View my Deep Learning Overview : [Google Slides] Deep Learning Research Projects: [Google Slides] Beware, these things get out of date very quick. This presentation is from Oct 2016. The outline of the talk: Toy Neural Network Loss Function Stochastic Gradient Descent Forward-pass (Neural Function Evaluation) Backward-pass (Gradient of Neural function wrt to params) Recent … Continue reading Deep Learning Overview →

Toy Neural Network

In my last post on neural network [HERE], I talked on how one can think of neural network as universal approximators. In this post I am trying to help understand a toy neural network implementation. In particular one can have a clearer and intuitive understanding of what a forward_pass is and what back_propagation means. Most … Continue reading Toy Neural Network →

Neural Network as Universal Approximators : Intuitive Explaination

Came across this wonderful explanation of why the neural network with hidden layer are universal approximators. Although not very helpful for practical purpose gives an intuitive feel of why neural network give reasonable results. The basic idea is to analyze a sigmoid function as you change w and b . In particular effect on $latex \sigma( w\times x … Continue reading Neural Network as Universal Approximators : Intuitive Explaination →

Graph Segmentation of Images

Use of graph representation of an image for segmentation. This is based on the following paper which is one of the most cited papers in Computer Vision. If you are starting to do research in computer vision related fields it is a good idea to understand few of those papers in as much detail as possible. … Continue reading Graph Segmentation of Images →

Tutorial Slides on Computer Network

I was a TA for HKUST course ELEC 4120 during spring of 2014 and spring of 2015. The slides here are supplementary material for the course. It contains additional explanations of the concepts, more numerical problems, and exercises. For more details of the concepts refer to the course notes of the instructor. Additionally and ideally … Continue reading Tutorial Slides on Computer Network →

Details of Intra-Prediction in High Efficiency Video Coding (HEVC)

The new High Efficiency Video Coding (HEVC) standard has been recently developed by the Joint Collaborative Team on Video Coding (JCT-VC) which was established by the ISO/IEC Moving Picture Experts Group (MPEG) and ITU-T Video Coding Experts Group (VCEG). This new standard will replace the current H.264/AVC standard to deal with nowadays and future multimedia market trends. This post … Continue reading Details of Intra-Prediction in High Efficiency Video Coding (HEVC) →

Gabor Image Features

Computation of Gabor Features - Mean Squared Energy, Mean Amplitude. In applications of computer vision and image analysis, Gabor ﬁlters have maintained their popularity in feature extraction for almost three decades. The original reason that draw attention was the similarity between Gabor ﬁlters and the receptive ﬁeld of simple cells in the visual cortex. A more practical reason … Continue reading Gabor Image Features →

Machine Learning : Handling Imbalanced Datasets

When dealing with real datasets in machine learning or data mining, we quite frequently encounter a 2 category classification task. However, to add to our agony the dataset is skewed. This means samples from one class are more in number than the other. There are a few well know techniques to get around the problem. … Continue reading Machine Learning : Handling Imbalanced Datasets →

Computing Padé Approximation with Maxima

Computationally intensive software programs can have a sharp performance profile. What I mean with sharp profile is that, there might be a couple of functions which are most time consuming. Very often such expensive functions are trigonometric functions. A way to increase performance is by use of approximations for trigonometric function. There is a trade … Continue reading Computing Padé Approximation with Maxima →