Back

Applications of NL(X) and LLM

95-820

Units: 6

Description

The rapid adoption of Generative AI technologies is reshaping industries across the globe, from healthcare to financial services. According to Bain & Company, 87% of companies were already developing, piloting, or deploying generative AI by the start of 2024. Moreover, McKinsey's research estimates that generative AI could contribute an additional $2.6 trillion to $4.4 trillion annually across various sectors. These figures underscore the urgent need for professionals equipped with specialized knowledge in Natural Language Processing and Understanding (NL(X)) and Large Language Models (LLMs).

This course is designed to provide graduate-level students with a comprehensive understanding of NL(X) and LLMs, focusing on their applications, evaluation, and operationalization across diverse industries. Beginning with the fundamentals of NL(X), the course covers its history, evolution, and critical applications, offering students hands-on experience with essential tools such as text mining, sentiment analysis, and embeddings.

As students’ progress, they will delve into advanced architectures, including RNNs, LSTMs, and Transformers, learning how these models drive key applications like machine translation and named entity recognition (NER). The course places significant emphasis on Large Language Models, such as GPT and BERT, guiding students through the intricacies of training, fine-tuning, and deploying these models. Advanced topics like Retrieval-Augmented Generation (RAG) and agentic architectures will also be explored, highlighting how LLM-based agents are transforming tasks that require complex reasoning, planning, and execution.

To bridge the gap between theory and practice, the course offers detailed instruction on LLMOps, covering best practices for transitioning models from development to production. This includes a strong focus on ethical considerations, operational risks, and the optimization of model performance in enterprise settings.

Students will also benefit from guest lectures by industry professionals, providing valuable insights into the practical challenges and opportunities of applying NL(X) and LLMs in real-world environments. These sessions are designed to help students connect their academic learning with industry needs, preparing them to lead in the fast-evolving field of AI.

Given the accelerating adoption of Generative AI, this course is essential for those aspiring to roles as NLP engineers, data scientists, AI analysts, or professionals looking to leverage LLMs to drive innovation. By the end of the course, students will possess the critical skills and knowledge required to develop, evaluate, and deploy advanced NL(X) and LLM solutions, positioning themselves at the forefront of AI-driven transformation.

Learning Outcomes

  • Understand the Evolution and Foundations of NLP: Students will trace the historical milestones in NLP, appreciate its significance in the modern world, and grasp foundational concepts including neural networks and their architectures.
  • Master Practical NLP Techniques and Tools: Students will be proficient in text mining techniques, understand popular NLP tools, analyze deep learning architectures (e.g., RNN, LSTM, Transformer) and apply advanced techniques like attention mechanisms to real-world scenarios. 
  • Develop and implement NLP algorithms for specific industries: Students will be able to develop and implement NLP algorithms for specific industries, such as financial services, healthcare, telecommunications, media, technology, retail, and manufacturing, to solve real-world problems.
  • Evaluate the capabilities and limitations of large language models: Students will be able to evaluate the limitations and capabilities of large language models, including their ability to develop emergent abilities and their potential for propagating errors or biases.
  • Fine-tune and adapt large language models for specific generative AI tasks: Students will be able to fine-tune and adapt large language models, such as GPT, LLaMA, PaLM for specific generative AI tasks, including document retrieval, Q&A, language generation, etc.
  • Implement generative agent models for reasoning, planning, and optimization: Students will be able to implement generative agent models for reasoning, planning, and optimization using large language models and agents to solve specific problems in various application areas, such as social analytics, report generation, chatbots, retrieval augmented generation, document processing, topic modeling, and named entity recognition.

Prerequisites Description

This course requires a basic background in data science and/or Artificial Intelligence. Basic level of Python programming is required for completing the assignments.

 

90803 or 17644 or 10601 or 95828

Syllabus