Linear equations and solutions Matrices and their Properties; Eigenvalues and eigenvectors; Matrix Factorizations; Inner products; Distance measures; Projections; Notion of hyperplanes; halfplanes.
Probability theory and axioms; Random variables; Probability distributions and density functions ;Expectations and moments; Covariance and correlation; Statistics and sampling distributions; Hypothesis testing of means, proportions, variances and correlations; Confidence intervals; Correlation functions; Parameter estimation – MLE and Bayesian methods
Unconstrained optimization; Necessary and Sufficiency conditions for optima; Gradient descent methods; constrained optimization, KKTConditions; Introduction to least squares optimization;
1. Use cases from the healthcare domain where NLP is applied
2. Models such as Bi-LSTM-CRF, CAML, HAN, ResNexT.
3. Public domain datasets - MIMIC-III.
Introduction to big data in biology
Levels of omics data, basic information flow in biology
Importance of Networks in Biology: Overview
Introduction to Network Science
Learning from Network structure: Predicting essential genes
Learning on Networks: Community detection to identify disease genes - Learning using Networks: Graph mining for predicting biosynthesis routes - Omic data analysis: Predicting mutations and genes that drive cancer
1. Problem Statement : Four case studies will be demonstrated. CS1: Choice of mode CS2: Travel time estimation CS3: Accident hot spot analysis CS4: Accident severity modelling
2. Model(s) intended to demonstrate : Logistics regression, Support vector regression, k-means clustering and random forest
3. Dataset to be used during the demo
4. Dataset for the mini project
1. Levels of omics data, basic information flow in biology
2. Genomics, Transcriptomics, Epigenomics, Proteomics and Multi omics - Identification human disease genes using genomics
3. Application of transcriptomics for identifying disease mechanisms
4. Clinical data - kinds of clinical data Garbhini dataset - a clinical data case study
a. Artificial Neuron
b. Multilayer Perceptron
c. Universal Approximation Theorem
d. Backpropagation in MLPs
e. Backprop on general graphs
a. Gradient Descent and its variants
b. Momentum, Adam, etc.
c. Batch Normalization
a. Introduction
b. CNN Operations
c. CNN Training
d. Illustrative Example (“Hello World”) - MNIST digit classification e. Image Recognition-SoTA model(s)
f. Object detection/localization - SoTA model(s)
g. Semantic segmentation -SoTA model(s)
a. Smart Cities
b. Industry Use case 1
c. Climate Science
d. Manufacturing
e. Bio-informatics
Industry Use case 2
**Dates will be decided keeping the safety of participants in mind. Fees will be based on actuals.