Trends in AI and Deep Learning Research Seminar (Sommersemester 2024)
Dozent:
Prof. Dr. Gerard de Melo
(Artificial Intelligence and Intelligent Systems)
,
jingyi Zhang
(Artificial Intelligence and Intelligent Systems)
Allgemeine Information
- Semesterwochenstunden: 4
- ECTS: 6
- Benotet:
Ja
- Einschreibefrist: 01.04.2024-30.04.2024
- Lehrform: Vorlesung / Übung
- Belegungsart: Wahlpflichtmodul
- Lehrsprache: Deutsch
Studiengänge, Modulgruppen & Module
- BPET: Business Process & Enterprise Technologies
- HPI-BPET-K Konzepte und Methoden
- BPET: Business Process & Enterprise Technologies
- HPI-BPET-T Techniken und Werkzeuge
- BPET: Business Process & Enterprise Technologies
- HPI-BPET-S Spezialisierung
- OSIS: Operating Systems & Information Systems Technology
- HPI-OSIS-K Konzepte und Methoden
- OSIS: Operating Systems & Information Systems Technology
- HPI-OSIS-T Techniken und Werkzeuge
- OSIS: Operating Systems & Information Systems Technology
- HPI-OSIS-S Spezialisierung
- DANA: Data Analytics
- HPI-DANA-K Konzepte und Methoden
- DANA: Data Analytics
- HPI-DANA-T Techniken und Werkzeuge
- DANA: Data Analytics
- HPI-DANA-S Spezialisierung
- CODS: Complex Data Systems
- HPI-CODS-K Konzepte und Methoden
- CODS: Complex Data Systems
- HPI-CODS-T Techniken und Werkzeuge
- CODS: Complex Data Systems
- HPI-CODS-S Spezialisierung
- Digital Health
- HPI-DH-DS Data Science for Digital Health
- DICR: Digitalization of Clinical and Research Processes
- HPI-DICR-C Concepts and Methods
- DICR: Digitalization of Clinical and Research Processes
- HPI-DICR-T Technologies and Tools
- DICR: Digitalization of Clinical and Research Processes
- HPI-DICR-S Specialization
- APAD: Acquisition, Processing and Analysis of Health Data
- HPI-APAD-C Concepts and Methods
- APAD: Acquisition, Processing and Analysis of Health Data
- HPI-APAD-T Technologies and Tools
- APAD: Acquisition, Processing and Analysis of Health Data
- HPI-APAD-S Specialization
- SECA: Security Analytics
- HPI-SECA-K Konzepte und Methoden
- SECA: Security Analytics
- HPI-SECA-T Techniken und Werkzeuge
- SECA: Security Analytics
- HPI-SECA-S Spezialisierung
- DSYS: Data-Driven Systems
- HPI-DSYS-C Concepts and Methods
- DSYS: Data-Driven Systems
- HPI-DSYS-T Technologies and Tools
- DSYS: Data-Driven Systems
- HPI-DSYS-S Specialization
- MALA: Machine Learning and Analytics
- HPI-MALA-C Concepts and Methods
- MALA: Machine Learning and Analytics
- HPI-MALA-T Technologies and Tools
- MALA: Machine Learning and Analytics
- HPI-MALA-S Specialization
Beschreibung
Deep Learning is the foundation for most modern approaches to AI, especially for large language models (LLMs), natural language processing (NLP), and computer vision (CV). Recent advances such as ChatGPT and other Transformer models as well as multimodal models such as Midjourney and CLIP show the effectivess of deep learning solutions in tackling many complex single-modality and multimodal problems.
This seminar aims to build upon recent research in LLMs, NLP, CV, and deep learning.
Potential topics that could be explored in this seminar include Transformer models for NLP, vision-and-language models, etc. The specific list of topics will be presented in the first session.
Voraussetzungen
The main focus is on research, so depending on the topic, some prior familiarity with ML/AI, especially Deep Learning, is probably needed. For many topics, you will need some prior experience with PyTorch, Tensorflow, or Jax, and prior experience in training deep neural networks with GPUs. For example, you can take our "Natural Language Processing" course, Christoph Lippert's Deep Learning course, or Dagmar Kainmüller's computer vision course to acquire the prerequisite knowledge. The specific prequisites for each topic will be explained in the first session.
Lern- und Lehrformen
This seminar focuses on practical research skills. Depending on the topic, you can either investigate it alone or in a team. Students will work on these projects throughout the semester, supported by weekly meetings with their mentor.
Leistungserfassung
The grade will be based on the following:
- 25% Final Presentation
- 75% Project (7 to 10-Page Paper and Code Submission)
Important criteria for the evaluation of the project include the project effort, the quality of the paper, and the reproducibility of the code. Further details will be given during the seminar.
Termine
Potential topics are presented on the first day (Monday, April 8, 15:15-16:45 in HPI Main Building, room H2.57/58).
There will be a mid-term presentation during the semester in addition to a final presentation at the end of the semester.
Slides for organization are here https://drive.google.com/file/d/106uJItKJoZzlE9q73qhfNqonWM7Su-Px/view?usp=sharing
Slides for the topics are here https://drive.google.com/file/d/13vD9qkz11TwaFgEAJ_ZWh-p3SKJ3COgd/view?usp=sharing
Zurück