Modules

Prerequisites

Contents

The module provides a comprehensive overview of the field of Conversational AI. • History of natural language processing (NLP) and chatbots, including the meaning of NLP, NLU, • Introduction to text mining using Matlab/Simulink, Python, R • Large language models (LLM) and multimodal models • Transformer models for deep learning (DL), including generative pre-trained transformer (GPT-3), GPT-3.5, bidirectional encoder representations from transformers (BERT), Turing natural language generation (Turing-NLG), and Wu Dao 2.0 • Creating and managing a conversational agent with Google Dialogflow CX including flows, pages, intents, entities, fulfillments, and webhooks) • Cybersecurity aspects of chatbots, misinformation, and data biases

Learning outcomes

After successfully completing the module, students have a comprehensive overview of the field of Conversational AI. Studnets know the history of NLP and chatbots, including the meaning of NLP, NLU, and NLG. They know text mining techniques using Matlab/Simulink, Python and R learn. They also have knowledge of large language models and multimodal models, and familiarity with Transformer models for deep learning. They know the limitations of models and new platforms such as ChatGPT, Baidu's Ernie, and Google's Bard and are able to build and maintain conversation-based agents.

Assessment

• Seminar presentation, 5-10 minutes (20%) • Paper/Essay, 10-15 pages (30%) • Successful completion of bi-weekly exercises (50%)

Remarks

• Study programme Artificial Intelligence M.Sc.: Compulsory elective module in complex „Seminars or Laboratories“ • Study programme Informatik M.Sc.: Compulsory elective module in complex „Seminare oder Praktika“ (level 400)