Below are some concepts and techniques which I found interesting. These include some advanced and interesting topics to explore in ML including - Time series Analysis, NLP, Computer Vision etc.

• “Grounding Dino” is a zero-shot learning model. Something like Image transformers. It does not require labeling while training. Link: Grounding DINO : SOTA Zero-Shot Object Detection

• Autoencoders can be used for image noise filtering and anomaly detection from data. Link: 85a - What are Autoencoders and what are they used for?

• For Anomaly Detection: 180 - LSTM Autoencoder for anomaly detection

• For tracking objects and counting unique objects Deep SORT is used. Link: Deep SORT

• BERT and GPT. : 308 - An introduction to language models with focus on GPT

• GPT is mainly focused on the “Decoding” part.

• BERT is focused on the “Encoding” part.

• Important steps regarding tflite : TensorFlow Lite | ML for Mobile and Edge Devices

• Take a look into YOLO-NAS : How to Train YOLO-NAS on a Custom Dataset

• Snapshot Ensemble Deep Learning: Snapshot Ensemble Deep Learning Neural Network in Python - MachineLearningMastery.com

• Selective search and other search techniques in deep learning: Selective Search for Object Detection (C++ / Python) | LearnOpenCV

• High precision vs high recall: ChatGPT

• Working principle and comparison of object detection algorithms: R-CNN, Fast R-CNN, Faster R-CNN, YOLO — Object Detection Algorithms

• Graph convolutional network: Understanding Graph Convolutional Networks for Node Classification

• Understanding sub-word tokenization: Understanding sub word tokenization used for NLP