Personal

Two Years of Technical Fridays

A Year of Fridays

Smart India Hackathon 2018 grand finale

Technical Fridays

Site launch

Data Science

False positive paradox

Loss functions

Optimizers

Methods of Hyperparameter optimization

The Bayesian Thinking - III

The Bayesian Thinking - II

The Bayesian Thinking - I

Dropout: Prevent overfitting

How deep should neural nets be?

Don't use sigmoid: Neural Nets

Scaling vs Normalization

Ensembling is the key

Computational graphs: Backpropagation

Gradient descent: The core of neural networks

Gradient boosted trees: Better than random forest?

Linear algebra: The essence behind deep learning

Data Mining: Knowledge discovery in databases

Anscombe's Quartet

The Curse of Dimensionality

Dealing with categorical data

Regularization

Evaluation metrics for classification and False positives

Simplicity doesn't imply accuracy

p-Value

Out-liars

Correlation is not causation

Overfitting and Underfitting

Data leakage: A big problem

Simpson's paradox

Email spam filtering: Text analysis in R

Friendship paradox: facebook

Moneyball: Why no prediction can't be made for baseball champion

Moneyball: How linear regression changed baseball

Machine Learning

Loss vs Accuracy

Loss functions

Methods of Hyperparameter optimization

A visual introduction to eigenvectors and eigenvalues

Scaling vs Normalization

Ensembling is the key

Gradient boosted trees: Better than random forest?

Data Mining: Knowledge discovery in databases

The Curse of Dimensionality

Regularization

Simplicity doesn't imply accuracy

Overfitting and Underfitting

Email spam filtering: Text analysis in R

Moneyball: Why no prediction can't be made for baseball champion

Moneyball: How linear regression changed baseball

R

Simpson's paradox

Email spam filtering: Text analysis in R

Moneyball: Why no prediction can't be made for baseball champion

Moneyball: How linear regression changed baseball

Python

Transfer learning: How to build accurate models

Dynamic Programming

Anscombe's Quartet

Dealing with categorical data

Regularization

Some Prime Thoughts

Friendship paradox: facebook

Turtle in Python: A Traffic light

Algorithms

Dynamic Programming

Shortest Path: Dijkstra's Algorithm

Divide and Conquer

Greedy Algorithms

Structure of the web

Cryptography

Some Prime Thoughts

How secure are we?

Mathematics

False positive paradox

The Bayesian Thinking - III

The Bayesian Thinking - II

The Bayesian Thinking - I

A visual introduction to eigenvectors and eigenvalues

Linear algebra: The essence behind deep learning

Some Prime Thoughts

The Rule of 72: Mathematics in everyday life

Mathematics and Beauty

Visualization

Anscombe's Quartet

Deep Learning

Matrix Multiplication in CUDA

PyTorch Basic Tutorial

Introduction to Panoptic Segmentation: A Tutorial

Evaluation metrics for object detection and segmentation: mAP

Quick intro to Instance segmentation: Mask R-CNN

Quick intro to semantic segmentation: FCN, U-Net and DeepLab

Converting FC layers to CONV layers

Data augmentation

Generative Adversarial Networks variants: DCGAN, Pix2pix, CycleGAN

Layer-specific learning rates

Quick intro to Object detection: R-CNN, YOLO, and SSD

Attention

Backpropagation Through Time

Autoencoder: Downsampling and Upsampling

Weight initialization in neural nets

Image captioning using encoder-decoder

The gradient problem in RNN

Why Batch Normalization?

Filters in Convolutional Neural Networks

Loss vs Accuracy

Generative models and Generative Adversarial Networks

Skip connections and Residual blocks

Loss functions

Optimizers

Transfer learning: How to build accurate models

Methods of Hyperparameter optimization

word2vec: The foundation of NLP

Dropout: Prevent overfitting

How deep should neural nets be?

Don't use sigmoid: Neural Nets

The magic behind ConvNets

Computational graphs: Backpropagation

Gradient descent: The core of neural networks

Linear algebra: The essence behind deep learning

Computer Vision

PyTorch Basic Tutorial

Color and color spaces in Computer Vision

Introduction to Panoptic Segmentation: A Tutorial

Evaluation metrics for object detection and segmentation: mAP

Quick intro to Instance segmentation: Mask R-CNN

Quick intro to semantic segmentation: FCN, U-Net and DeepLab

Converting FC layers to CONV layers

Data augmentation

Generative Adversarial Networks variants: DCGAN, Pix2pix, CycleGAN

Quick intro to Object detection: R-CNN, YOLO, and SSD

Attention

Image captioning using encoder-decoder

Why Batch Normalization?

Filters in Convolutional Neural Networks

Generative models and Generative Adversarial Networks

Skip connections and Residual blocks

Transfer learning: How to build accurate models

The magic behind ConvNets

Natural Language Processing

Retrieval Augmented Generation (RAG) Chatbot for 10Q Financial Reports

Attention

Backpropagation Through Time

Image captioning using encoder-decoder

The gradient problem in RNN

word2vec: The foundation of NLP

Generative AI

Matrix Multiplication in CUDA

Retrieval Augmented Generation (RAG) Chatbot for 10Q Financial Reports

Generative Adversarial Networks variants: DCGAN, Pix2pix, CycleGAN

Generative models and Generative Adversarial Networks

Speech Recognition

Introduction to Automatic Speech Recognition

PyTorch

PyTorch Basic Tutorial

LLM

Matrix Multiplication in CUDA

Retrieval Augmented Generation (RAG) Chatbot for 10Q Financial Reports

CUDA

Matrix Multiplication in CUDA

Optimization

Matrix Multiplication in CUDA