Doctoral Consortium at ECAI-2023
Doctoral Consortium is a full-day workshop that will take place on Saturday, September, 30, 2023. It includes short presentations, an invited talk, a career panel discussion and a poster session.
The event will take place in the building of the Faculty of Physics, Astronomy and Applied Computer Science in room A-1-06.
Please see below a preliminary schedule.
|09:00 – 09:15||Opening/Welcome|
|09:15 – 10:30||Student talks (8 papers) + Elevator pitches (5 posters)|
|10:30 – 11:00||Coffee Break|
|11:00 – 11:30||Posters|
|11:30 – 12:30||Invited talk (by Ulle Endriss)|
|12:30 – 13:30||Lunch Break|
|13:30 – 15:00||Student talks (10 papers) + Elevator pitches (4 posters)|
|15:00 – 15:30||Coffee Break|
|15:30 – 16:00||Posters|
|16:00 – 17:00||Career Panel|
|17:00 – 17:15||Closing remarks|
Instructions for Students
Each presentation will be 6 minutes + 2 minutes for questions. Please prepare 5 slides (including the title slide). Please make your presentation accessible to broad computer science audience. Each student who will give an oral presentation can additionally prepare a poster. The posters will be presented at the dedicated sessions. Some of the submissions were accepted only as posters. For such submission the presenting student should additionally prepare a 90 seconds spotlight talk advertising the poster, using at most 2 slides.
The posters should be A0 Portrait size. Ideally, the posters should be made of a lightweight material (for example, paper or a very light textile).
Please send your slides to us by September 22 so we can compile one presentation desk per session to avoid any setup delays.
Please let us know if you have any hard time constraints, e.g. you have a presentation at a different workshop/tutorial by September 15th.
Speaker: Ulle Endriss, University of Amsterdam
Title: How to write a review [slides for the lecture]
Abstract: During this talk we will be discussing best practices for writing reviews for papers submitted to AI conferences and journals. We also will be touching on difficult questions such as these: Who should write the reviews? Should they be paid for their work? Should the reviewing process be anonymous and what does that entail?
About the speaker: Ulle Endriss is Professor of AI and Collective Decision Making at the University of Amsterdam. He is a EurAI Fellow and best known for his contributions to computational social choice. Closely related to the topic of this talk, he has served on the programme committees of well over 100 conferences and workshops, and he has been Associate Editor of both AIJ and JAIR. He was PC chair of AAMAS-2021 and will be PC chair of ECAI-2024.
During the career panel the students will be able to get advice on managing their career. We will host four excellent and experienced researchers as panelists:
Students Presenting in the Morning
- Concepts, Relations and Rules Extraction from Trained Convolutional Neural Networks: A Framework for Explainability.
Eric Ferreira dos Santos
- Theory of Attention Networks in Deep Learning.
- Multi-Source Domain Adaptation through Wasserstein Barycenters.
Eduardo Fernandes Montesuma
- Neural Networks based on Differential Equations for Modelling Real Systems.
- A Unified Framework for Reproducibility in Deep Learning.
- Implicitly Cooperative Agents through Impact-Aware Learning.
- Towards responsible and explainable deep learning on X-Ray and CT imaging.
- Translation Quality in Translation Studies and Computational Linguistics/Machine Translation.
- Goal-oriented ML methods to meet stakeholder requirements in manufacturing use cases.
- Monte Carlo tree search with state merging for reinforcement learning in Regular Decision Processes.
Gabriel Paludo Licks
- Explainable Artificial Intelligence through agents argumentation and abstractions.
- Explainable Deep Reinforcement Learning through Desired Goals and Predicted Paths.
Students Presenting in the Afternoon
- Trustworthy Autonomous Systems Through Social Explainable AI.
- Causal Discovery and Argumentation for Transparent Knowledge Discovery.
- Self-supervision and Controlling Techniques to Improve Counter Speech Generation.
- Fair Algorithms for Machine Learning.
- Uncovering Biases in Face Recognition Systems.
- Decentralized Asynchronous Multi-Agent Active Search.
- Plan Recognition.
- Action-Failure Resilient Planning.
- Methods for Diagnosis of Multi-Agent Systems.
- On AI and Education – Integrating AI Competences in Engineering Education.
- Learning from partial observations with RDPs.
- Moral Planning Agents.
- A Strategic Approach to Deceptive Planning in Multi-Agent Simulations.
- Training Reinforcement Learning Model Based on Natural Language Understanding.
- Attribution Explanations for Quantitative Argumentation Frameworks.