Writings

Essays and articles on data science, generative AI, and applied machine learning.

Transformers: Attention Mechanism

Transformers: Attention Mechanism

Transformers Attention Mechanism The breakthrough that allows models to focus on relevant parts of the input at each generation step. Introduction The previous article introduced the encoder-decoder …

machine-learningdata-scienceaideep-learning
Transformers: Encoder-Decoder Architecture

Transformers: Encoder-Decoder Architecture

Transformers Encoder-Decoder Architecture The foundational blueprint behind machine translation, summarization, and sequence-to-sequence tasks. Introduction If you have ever used Google Translate, a …

transformersmachine-learningaideep-learning
Natural Language Processing: Text Representations

Natural Language Processing: Text Representations

Natural Language Processing Text Representations From Bag of Words to TF-IDF, the foundational methods for converting text into numbers. Introduction There is a fundamental challenge in machine …

deep-learningnlpdata-scienceai
Sequential Data Modeling: Text Generation with RNNs

Sequential Data Modeling: Text Generation with RNNs

Sequential Data Modeling Text Generation with RNNs Training neural networks to generate text character by character applied to Shakespeare and weather forecasting. Introduction This article puts …

data-sciencemachine-learningdeep-learningtext-generation
Natural Language Processing: Real World NLP with Word2Vec

Natural Language Processing: Real World NLP with Word2Vec

Natural Language Processing Real World NLP with Word2Vec From toy examples to a complete document classification pipeline using pre-trained embeddings. Introduction The previous article trained …

word2vecmachine-learningdeep-learningnlp
Deep Learning Training Enhancements: Residual Connections and ResNets

Deep Learning Training Enhancements: Residual Connections and ResNets

Deep Learning Training Enhancements Residual Connections and ResNets The skip connections that made training 100+ layer networks possible. Introduction Before residual connections were introduced …

deep-learningdata-sciencemachine-learningai
Sequential Data Modeling: LSTM and GRU

Sequential Data Modeling: LSTM and GRU

Sequential Data Modeling LSTM and GRU The gating mechanisms that solved the vanishing gradient problem in recurrent networks. Introduction Standard RNNs have a fundamental limitation: they struggle …

machine-learninglstmdata-sciencedeep-learning
Natural Language Processing: Word Embeddings and Word2Vec

Natural Language Processing: Word Embeddings and Word2Vec

Natural Language Processing Word Embeddings and Word2Vec Dense vector representations that capture the meaning of words. Introduction Word embeddings are the foundation of virtually every modern NLP …

word-embeddingsdata-sciencenlpmachine-learning
Deep Learning Training Enhancements: Batch Normalization

Deep Learning Training Enhancements: Batch Normalization

Deep Learning Training Enhancements Batch Normalization The technique that made deep networks trainable and accelerated convergence. Introduction Before batch normalization was introduced in 2015 …

batch-normalizationmachine-learningdeep-learningdata-science
Sequential Data Modeling: Recurrent Neural Networks

Sequential Data Modeling: Recurrent Neural Networks

Sequential Data Modeling Recurrent Neural Networks How neural networks process sequential data by maintaining memory across time steps. Introduction The neural networks covered in previous articles …

data-sciencemachine-learningrnnnlp
More Articles On Medium