Softskills Seminar
This course is an obligatory course of the M2 of the Master's program “Data AI” of the Institut polytechnique de Paris - open to students of other programs as well. The purpose of this course is to train students to give scientific presentations.
Every student chooses one research paper from the list of proposed papers. The student then prepares a 20min presentation about this paper. For this purpose, she/he can request the help of the advisor of the paper (by email and/or by meeting with them). The student then gives the presentation in the allocated time slot of the Softskills seminar. Students are warmly encouraged to take into account the advice on giving good talks dispensed during the first session.
Each presentation is followed by a question-answer session, where both the students and the lecturers can ask the presenter questions about the paper. To animate this, each student is assigned to some other paper as the “devil's advocate”. In this role (which is not known to the other students), she or he prepares some questions for the presenter. However, all students are invited to participate in the question-answer session.
The course is graded with 80% presentation + 20% oral participation (including as devil's advocate).
Planning
The course takes place on Thursdays 13:30-16:30 -- always online, and sometimes in addition physically. There are 3 slots per session: 13:30, 14:30, 15:30 (allowing for the talk of 20 min, questions, and time to set up the remote video session).
10/9/2020: Amphi 0C02 at Télécom Paris
- Introduction to the Softskills Seminar (lecturer Fabian Suchanek)
- How to give good talks (lecturer Fabian Suchanek)
17/9/2020: Room 0A214 at Télécom Paris
- How to give interesting talks (lecturer Jean-Louis Dessalles)
- Introduction to research (lecturer Fabian Suchanek)
24/9/2020: Room 0A214 at Télécom Paris
- Pietro Gori 1: Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (Aleksa Marušić)
- David Filliat 2: World Models (Jérémy Perez)
- Filippo Miatto 1: Opening the Black Box of Deep Neural Networks via Information (Yazid Moulin)
1/10/2020: Room 0A214 at Télécom Paris
- Isabelle Bloch 2: GraphSleepNet: Adaptive Spatial-Temporal Graph Convolutional Networks for Sleep Stage Classification (Syrine El Aoud)
- Isabelle Bloch 3: Measurable Counterfactual Local Explanations for Any Classifier (Rodrigo Cézar LUZ BRO)
- Chloé Clavel 2: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (Alexandre Abou Chahin)
8/10/2020: Amphi 3 at Télécom Paris
- Natalia Diaz Rodriguez 2: Poincaré maps for analyzing complex hierarchies in single-cell data (Qi GAN)
- Louis Jachiet 1: VLog: A Rule Engine for Knowledge Graphs (Louis CALDAS)
- Winston Maxwell 1: Identifying the right level of explanation (Arnaud de Guilher)
15/10/2020: Amphi 3 at Télécom Paris
- Florence d'Alché 1: Optimal Transport for structured data with application on graphs (Amine Bejaoui)
- Florence Tupin 1: Occlusion Boundary Detection via Deep Exploration of Context (MAZEAU Brice)
- Florence Tupin 2: High-Quality Self-Supervised Deep Image Denoising (Dhif Ahmed)
22/10/2020: Amphi 3 at Télécom Paris
- Goran Frehse 1: Constrained Policy Optimization (EL KHALLIOUI AYOUB)
- Goran Frehse 3: Constrained Cross-Entropy Method for Safe Reinforcement Learning (Alex Vigneron)
- Fabian Suchanek 2: Robust Discovery of Positive and Negative Rules in Knowledge Bases (Alicia Breidenstein)
- Sophie Chabridon 1: Explainable Machine Learning in Deployment (Jean Cyrus de Gourcuff)
- Jean-Louis Dessalles 1: An impossibility theorem for clustering
- Jean-Louis Dessalles 2: One shot learning of simple visual concepts
- Jean-Louis Dessalles 3: The rational basis of representativeness
- Alexandre Chapoutot 1: Free-space Polygon Creation based on Occupancy Grid Maps for Trajectory Optimization Methods
- David Filliat 1: Neural SLAM: Learning to Explore with External Memory
- Isabelle Bloch 1: Hierarchical Attention Based Spatial-Temporal Graph-to-Sequence Learning forGrounded Video Description
- Isabelle Bloch 4: Classification Rules in Relaxed Logical Form
- Natalia Diaz Rodriguez 1: On the Benefits of Invariance in Neural Networks
- Natalia Diaz Rodriguez 3: Invariant Causal Prediction for Block MDPs
- Florence d'Alché 2: Convolutional Kernel Networks for Graph-Structured Data
- Angelos Anadiotis 1: Serverless in the Wild: Characterizing and Optimizing the Serverless Workload at a Large Cloud Provider
- Angelos Anadiotis 2: BitWeaving: fast scans for main memory data processing
- Goran Frehse 2: Algorithms for CVaR Optimization in MDPs
- Fabian Suchanek 1: Realistic Re-evaluation of Knowledge Graph Completion Methods: An Experimental Study
- Fabian Suchanek 3: DRUM: End-To-End Differentiable Rule Mining OnKnowledge Graphs
- Julien Alexandre dit Sandretto 1: Revising Hull and Box Consistency
- Chloé Clavel 1: Pretraining Methods for Dialog Context Representation Learning