Génie industriel - Rubrique Formation - 2022

Conversational generative AI for industry - 4GMC14C2

  • Number of hours

    • Lectures 6.0
    • Projects -
    • Tutorials 9.0
    • Internship -
    • Laboratory works -
    • Written tests 1.0

    ECTS

    ECTS 1.5

Goal(s)

This UE is an introduction to Conversational Generative AI. The objectives are:

  • Understand the main concepts, issues and applications of conversational generative AI in industrial engineering
  • Understand the main technical principles of conversational generative AI (natural language processing operations, architectures, LLM, prompting) and their limitations
  • Be able to adapt a generic LLM for an industrial engineering application
  • Be able to design and evaluate human-conversational generative AI interactions for industrial engineering tasks

Responsible(s)

Romain PINQUIE

Content(s)

Introduction, uses, and technologies:

  • 1CM: Learn about the challenges and applications of conversational generative AI in industrial engineering:
    -- What is a generative model?
    -- What are the use cases in industrial engineering?
    -- What are the main considerations? Cost, resources, infrastructure, bias, liability, etc.

    Understand the main principles of conversational generative AI (natural language processing operations, architectures, LLM, prompting) and their limitations:
  • 1CM: LLM + Architectures + Prompt engineering
    -- What is a Large Language Model?
    -- LLMs: BERT, GPT, Llama, Mistral, etc.
    -- Transformer
    -- Training + Fine-tuning (LoRA, QLoRA)
    -- Structure, zero-/one-/few-/multi-shot, chain-of-thought, prompt iteration
    -- ...
  • 1TP : NLP basics (tokenization, stemming, lematization, sentence splitting, NER, stop words removal, vector space model, bag of words, word embeddings)
  • 1 TP : Prompt engineering

    Be able to adapt a generic LLM for an industrial engineering application:
  • 1CM: tuning techniques: few shots learning, RAG, graphRAG, fine tuning, etc.
  • 3TP: application of tuning techniques

    Designing and evaluating human-IA conversational interactions for industrial engineering tasks:
  • 1CM: Design and evaluation of conversational generative human-IA interaction
  • 1TP: Evaluation of conversational generative human-IA interaction

Prerequisites

Students will have taken and validated the following courses: Probability and Statistics; Programming with R

Test

    The following weighting of grades is compatible with distance exams

    Session 1:
    - N1 = 0.5 x CC1 + 0.5 x ET1

    Session 2:
    - N2 = 0.5 x CC1 + 0.5 x ET2

    Cette pondération est compatible avec une organisation des enseignements et des examens en distanciel

    Session 1:
    - N1 = 0.5 x CC1 + 0.5 x ET1

    Session 2:
    - N2 = 0.5 x CC1 + 0.5 x ET2

    Calendar

    The course exists in the following branches:

    • Curriculum - Engineer student Master SCM - Semester 7
    • Curriculum - Engineer student Master PD - Semester 7
    see the course schedule for 2025-2026

    Additional Information

    Course ID : 4GMC14C2
    Course language(s): FR

    You can find this course among all other courses.

    Bibliography

    Manning, Christopher D., and Hinrich Schutze. 1999. Foundations of Statistical Natural Language Processing. The MIT Press. Cambridge, Mass.: MIT Press.

    Phoenix, J., & Taylor, M. (2024). Prompt Engineering for generative AI. "O'Reilly Media, Inc.".

    Raschka, S. (2024). Build a Large Language Model (From Scratch). Simon and Schuster.

    Alammar, J., & Grootendorst, M. (2024). Hands-on large language models: language understanding and generation. " O'Reilly Media, Inc.".

    Cila, N. (2022). Designing Human-Agent Collaborations: Commitment, responsiveness, and support. CHI Conference on Human Factors in Computing Systems, 1–18. https://doi.org/10.1145/3491102.3517500

    Amershi, S., Weld, D., Vorvoreanu, M., Fourney, A., Nushi, B., Collisson, P., Suh, J., Iqbal, S., Bennett, P. N., Inkpen, K., Teevan, J., Kikin-Gil, R., & Horvitz, E. (2019). Guidelines for Human-AI Interaction. Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, 1–13. https://doi.org/10.1145/3290605.3300233

    White, J., Fu, Q., Hays, S., Sandborn, M., Olea, C., Gilbert, H., Elnashar, A., Spencer-Smith, J., & Schmidt, D. C. (2023). A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT (No. arXiv:2302.11382). arXiv. http://arxiv.org/abs/2302.11382

    French State controlled diploma conferring a Master's degree

    diplôme conférant grade de master contrôlé par l'Etat
    Common Core
    Common Core presentation
    Programme courses S5
    Programme courses S6
    Supply Chain Management
    Programme presentation
    Programme courses S7
    Programme courses S8
    Programme courses S9
    Programme courses S10
    Product Design
    Programme presentation
    Programme courses S7
    Programme courses S8
    Programme courses S9
    Programme courses S10

    Contacts

    Academic staff

    Registrar's office