Nilesh BarlaA short note on Reinforced Fine-Tuning or ReFTLLMs still lack strong reasoning capabilities. How can Reinforced Fine-Tuning (ReFT) can solve it?Sep 3Sep 3
Nilesh BarlaAdvice on learning advanced deep learning codingThe most effective way to learn to code is by studying the entire codebase and implementing it step by step.Jul 15Jul 15
Nilesh BarlainPerceptronAIA study on Attention mechanismThis study shows the impact, evolution, and performance of the attention mechanisms. The article highlights the mathematical approach to…Jul 14Jul 14
Nilesh BarlaEnhancing Language Models through Retrieval Augmented GenerationRetrieval Augmented Generation (RAG) emerges as a promising approach to overcome challenges faced by Language Model (LLM) systems. LLMs…Mar 20Mar 20
Nilesh BarlainPerceptronAIReview: Spectral normalization for GANsGenerative adversarial networks or GANs are one of the most popular methods to generate samples. They are implicit models which means that…Nov 24, 2021Nov 24, 2021
Nilesh BarlainPerceptronAIAdvances in Natural Language ProcessingAn overview of how NLP evolved over the yearsMar 16, 20211Mar 16, 20211
Nilesh BarlainPerceptronAIFine Tuning BERT: Multilabel Text ClassificationIt is not news for the NLP community that the state-of-the-art pre-trained language model BERT (Bidirectional Encoder Representations from…Mar 16, 2021Mar 16, 2021
Nilesh BarlainAnalytics VidhyaDetecting Outliers and its TreatmentExperiments and observationsMar 13, 20201Mar 13, 20201
Nilesh BarlainAnalytics VidhyaStyle Transfer with Deep LearningImplementation with PytorchNov 15, 2019Nov 15, 2019
Nilesh BarlainAnalytics VidhyaUnderstanding Gradient DescentImplementation with NumpyOct 24, 2019Oct 24, 2019
