Summary of data science/ML experience from online materials

• Data Science and Machine Learning

  • An introduction to Statistical Learning with Application in Python including Linear Regression, Logistic Regression, SVM, Decision Trees, Unsupervised Learning with PCA and k-Means clustering
  • Machine Learning Specialization by Andrew NG covering supervised, unsupervised and reinforcement learning.

• Natural Language Processing

  • NLP with machine learning (CS224N) exercise using neural networks, LSTMs and transformer architecture for various NLP tasks.
  • LLM from stratch, engineered a chatbot from scratch and fine-tuned pre-trained GPT models for text identification.
  • Natural Languge Processing with Transformers Developed skills in building, debugging, and optimising transformer models for core NLP tasks such as text classification, named entity recognition, and question answering. Applied transformers for cross-lingual transfer learning and real-world scenarios with limited labeled data.
  • Knowledge graphs for RAG using Neo4j: Developed proficiency in using Neo4j's Cypher query language to manage and retrieve data from knowledge graphs. Built a question-answering system by integrating Neo4j with LangChain, enhancing the performance of large language models (LLMs) with structured, relevant context.

• Computer vision

  • Convolutional Neural Networks for Visual Recognition (CS231n): Gained expertise in implementing, training, and debugging convolutional neural networks (CNNs) for image classification and other vision tasks. Developed a deep understanding of cutting-edge research in computer vision and applied multi-million parameter networks to real-world problems.

• Reinforcement Learning

  • David Silver's RL course: covering key concepts such as Markov Decision Processes, dynamic programming, Monte Carlo methods, and temporal difference learning. Developed practical skills in implementing and optimising reinforcement learning algorithms for various applications, including game playing and robotic control. Gained a deep understanding of policy gradient methods and value function approximation.