Artificial Intelligence Projects

Q-Learning

Deep Q Learning

Deep Convolutional Q learning

-- Deep Q-Learning for Atari Breakout

Actor Critic Model

-- CartPole

A3c

DDPG

-- Inverted Pendulum control problem

Twin Delayed DDPG

ARS

NLP

Text classification with bag of words model built with RNN

-- Analyzing restaurant reviews positive or negative https://keras.io/examples/nlp/text_classification_from_scratch/

Sentiment analysis with NLTK

-- NLTK VADER sentiment analysis on imdb movie reviews

Topic modeling

-- Labeling quora questions

Text Generator

-- RNN based Text Generation model: Asking machine learning to write a bengali poem

chatbot (seq2seq model)

https://github.com/samurainote/Automatic-Encoder-Decoder_Seq2Seq_Chatbot

Machine Translation

https://github.com/samurainote/seq2seq_translate_slackbot https://github.com/nrkapri/practical_seq2seq https://github.com/nrkapri/seq2seq-chatbot

Question Answering chatbot Squad

Abstract Text Summarization

https://github.com/samurainote/Text_Summarization_using_Bidirectional_LSTM

BERT

HuggingFace Megatron RoBerta Transformers Glove word2vec gensim

ELMo

generating image description

Metrics ROUGE, BLEU

chitchat bot question

Natural Language Generation MLT Summarization:

Chit Chat -- Dialog Creative Writing -- story telling, poetry generation Freeform Question answering Image Captioning

NeuralGen Workshop Storytelling Workshop Alexa Challenge ConvAI2 NeuralPS

Decoding Algo: Beam Search Greedy

CNN for NLP: Question

Question Answering over Text: MC Test Reading Comprehension Test Squad DeepQA Jeopardy IBM

Standford Attention reader FusionNet

Named Entity Recognition

GPT-3 https://keras.io/examples/nlp/

Project structure :

1. Abstract Introduction
2. Prior related work
3. Model
4. Model
5. Data
6. Experiments
7. Results
8. Analysis and Conclusion

Machine learning Projects

Regression

-- Intuition

-- Predict Wine Quality score

Classification

-- Intuition

-- Multiclass classification with Iris Dataset.

Clustering

-- Intuition

K-means clustering

-- K-Means Clustering to identify customer segments

hierarchical clustering

-- Hierarchical Clustering to identify customer segments

Association Rule :apriori

-- Intuition

-- cart optimization

Dimensionality reduction

Principal component analysys

-- Dimentionality Reduction with PCA

Linear Discriminant Ananlysys

-- Dimentionality Reduction with LDA

Kernel PCA

-- Dimentionality Reduction with Kernel PCA

Model Selection

-- kfold-cross-validation

-- Hyperparameter tuning with Grid Search

XGBoost

-- XgBoost Demo

Deep Learning

Supervised learning

ANN

-- Classifying malignant/Benign cancer with ANN

CNN

-- Image classifier for Fashion MNIST dataset

-- CNN deep dream visualization

-- Emnist digit classification visualizer

-- What Neural Networks See

-- Neural Style transfer

RNN

-- Predicting Google Stock price with RNN

music composer

Unsupervised learning

SOM

-- Detecting Fraud in Credit Card Applications

Boltzman machine

Autoencoder

-- Noise removal from Emnist image data with AutoEncoders

Original	Noisy	Reconstructed

GAN

-- Deep Fake video generation

Object Detection

-- Object detection with facebook's Detectron2: Inference with pre-trained model

-- Object Detection with Tensorflow FasterRCNN+InceptionResNet V2

Tools

1. python
   • https://www.geeksforgeeks.org/python-programming-language/
   • https://github.com/nrkapri/machine-learning/blob/master/tools/GeeksforGeeks_Python_tutorial.ipynb
2. numpy
3. pandas
4. matplotlib and seaborn
5. tensorflow and keras
   • https://www.tensorflow.org/tutorials/
   • https://keras.io/examples/
6. pytorch
   • https://pytorch.org/tutorials/