Differences

This shows you the differences between two versions of the page.

--- curriculum [2024/07/05 13:08] – 84.14.146.134
+++ curriculum [2025/01/10 12:52] (current) – [Period 2] respai-vic
@@ Line 1: / Line 1: @@
-===== Detailed Curriculum of the MScT AI-VIC=====
+===== Detailed Curriculum =====
 ===== MASTER 1  (M1) =====
@@ Line 12: / Line 12: @@
 Mandatory:
-  * Computer Animation (INF585, Marie-Paule Cani, EP)
+  * Computer Animation (INF585, Mathieu Desbrun and Marie-Paule Cani, EP)
 course among:
@@ Line 42: / Line 42: @@
   * Image Synthesis: Theory and Practice (INF584, Tamy Boubekeur, Telecom ParisTech)
   * Real-time AI in Video Games : Decisive & Collaborative Actions (INF584A, David Bilemdjian, Chaire Ubisoft)
-  * Statistics in Action (MAP566, Julien Chiquet, EP & Agro Paris Tech)
+  * Statistics in Action (MAP566, Zacharie Naulet, EP & INRAE)
   * Computational Optimal Transport for ML and Generative Modeling (MAP588, Rémi Flamary, EP)
@@ Line 62: / Line 62: @@
   * MAP630 - Refresher in Statistics: statistical analysis, introduction to Machine Learning techniques (Pierre Latouche, CNRS).
-  * INF630 - Refresher in Computer Science: Algorithmic geometry, Computer Graphics (Pooran Memari, CNRS) and Character animation on Unity  (Marie-Paule Cani, EP).
+  * INF630 - Refresher in Computer Science: Algorithmic geometry (Pooran Memari, CNRS), Computer Graphics and Character animation on Unity (Marie-Paule Cani, EP).
 ==== Period 1 ====
-**MAP/INF631 - Deep Learning (48h, 4 ECTS), Umut Simsekli (Inria), Pierre Latouche (CNRS)** (contact: umut.simsekli@inria.fr; pierre.latouche@uca.fr)
+**MAP/INF631 - Deep Learning (48h, 4 ECTS), Umut Simsekli (Inria), Pierre Latouche (UCA)** (contact: umut.simsekli@inria.fr; pierre.latouche@uca.fr)
-  Deep Learning is one key element of modern data science. This course will explore several instances of Deep Neural Networks, each one being specifically adapted to solve a particular learning task (classification, image recognition, text mining, dimensionality reduction). An introduction to current research topics on neural network will be presented during the last part of the course.
+> Deep Learning is one key element of modern data science. This course will explore several instances of Deep Neural Networks, each one being specifically adapted to solve a particular learning task (classification, image recognition, text mining, dimensionality reduction). An introduction to current research topics on neural network will be presented during the last part of the course.
 **INF632 - Introduction to NLP and LLMs (24h, 2 ECTS), Michalis Vazirgiannis (EP) ** (contact: michalis.vazirgiannis@polytechnique.edu)
-  This course offers an extensive and in-depth exploration of natural language processing (NLP) and large language models (LLMs), blending foundational principles with advanced techniques. The curriculum spans the domain of NLP, starting from basic concepts like indexing, Bag-of-Words, and TF-IDF, and extending to more sophisticated methods such as Graph of Words, Word2Vec, fastText, and GloVe. Then the focus shifts to deep learning models in NLP, including 1D-CNN, RNN, Hierarchical Attention Networks (HAN), and Neural Machine Translation (NMT).  Additionally, It delves into essential evaluation metrics and benchmark. Following that, the course presents the cutting-edge transformer architecture and pretraining techniques such as masked and causal language modelling. Key models including BERT, GPT, and BART are studied in detail, highlighting their transformative impact on the field and all their applications. The course also covers the emergence of LLMs introduced by ChatGPT, emphasizing prompting techniques, model adaptation using methods like LoRA, and model quantization techniques including techniques like QLoRA that merges both model adaptation and quantization. This comprehensive course integrates hands-on lab sessions using PyTorch and Keras, ensuring students gain practical experience alongside theoretical knowledge, preparing them for advanced roles in data science and NLP.
+> This course offers an extensive and in-depth exploration of natural language processing (NLP) and large language models (LLMs), blending foundational principles with advanced techniques. The curriculum spans the domain of NLP, starting from basic concepts like indexing, Bag-of-Words, and TF-IDF, and extending to more sophisticated methods such as Graph of Words, Word2Vec, fastText, and GloVe. Then the focus shifts to deep learning models in NLP, including 1D-CNN, RNN, Hierarchical Attention Networks (HAN), and Neural Machine Translation (NMT).  Additionally, It delves into essential evaluation metrics and benchmark. Following that, the course presents the cutting-edge transformer architecture and pretraining techniques such as masked and causal language modelling. Key models including BERT, GPT, and BART are studied in detail, highlighting their transformative impact on the field and all their applications. The course also covers the emergence of LLMs introduced by ChatGPT, emphasizing prompting techniques, model adaptation using methods like LoRA, and model quantization techniques including techniques like QLoRA that merges both model adaptation and quantization. This comprehensive course integrates hands-on lab sessions using PyTorch and Keras, ensuring students gain practical experience alongside theoretical knowledge, preparing them for advanced roles in data science and NLP.
 **INF639 - Deep Reinforcement Learning (24h, 2 ECTS), Jesse Read** (contact: jesse.read@polytechnique.edu)
-  Reinforcement learning (RL) is of increasing relevance today, including in games, complex energy systems, recommendation engines, finance, logistics, and for auto-tuning the parameters of other learning frameworks. This course assumes familiarity with the foundations of RL and its main paradigms (temporal-difference learning, Monte Carlo, and policy-gradient methods). We will explore them further, and study modern state-of-the-art variants (such as proximal policy optimization), with a focus on developing RL solutions with deep neural architectures suited to modern applications. We will also take a look at specialized topics such inverse reinforcement learning.
+> Reinforcement learning (RL) is of increasing relevance today, including in games, complex energy systems, recommendation engines, finance, logistics, and for auto-tuning the parameters of other learning frameworks. This course assumes familiarity with the foundations of RL and its main paradigms (temporal-difference learning, Monte Carlo, and policy-gradient methods). We will explore them further, and study modern state-of-the-art variants (such as proximal policy optimization), with a focus on developing RL solutions with deep neural architectures suited to modern applications. We will also take a look at specialized topics such inverse reinforcement learning.
 **INF631 - Analysis and Deep Learning on Geometric Data (24h, 2 ECTS), Maks Ovsjanikov (EP)** (contact: maks@lix.polytechnique.fr)
-  This course will introduce students to advanced topics in modern geometric data analysis with focus on a) mathematical foundations (discrete differential geometry, mapping, optimization), and b) deep learning for best performing methods. We will give an overview of the foundations in shape analysis and processing before moving to modern techniques based on deep learning for solving problems such as shape classification, correspondence, parametrization, etc.
+> This course will introduce students to advanced topics in modern geometric data analysis with focus on a) mathematical foundations (discrete differential geometry, mapping, optimization), and b) deep learning for best performing methods. We will give an overview of the foundations in shape analysis and processing before moving to modern techniques based on deep learning for solving problems such as shape classification, correspondence, parametrization, etc.
-**INF633 - Smart models for 3D Creation and Animation (24h, 2 ECTS),  Marie-Paule Cani (EP), Julien Pettré (Inria), Paul Boursin (EP), Antonin Wattel (Telecom)** (contact: marie-paule.cani@polytechnique.edu)
+**INF633 - Smart models for 3D contents Creation and Animation (24h, 2 ECTS),  Marie-Paule Cani (EP), Julien Pettré (Inria), Paul Boursin (EP), Antonin Wattel (Telecom)** (contact: marie-paule.cani@polytechnique.edu)
-  This course presents recent advances in 3D computer graphics, and more specifically in the subfields on modeling and animation, which rely on artificial intelligence. We first introduce user-centered Creative AI, i.e. smart 3D models - either based on knowledge or on deep learning from examples, designed to help users creating 3D virtual environments. Second, we focus the use of AI - from light models to deep reinforcement learning - in Character Animation, i.e. towards the training of autonomous 3D characters able to navigate and interact with such environments. The lab sessions are held on Unity, based on C#.
+> This course presents recent advances in 3D computer graphics, and more specifically in the subfields on modeling and animation, which rely on artificial intelligence. We first introduce user-centered Creative AI, i.e. smart 3D models - either based on knowledge or on deep learning from examples, designed to help users creating 3D virtual environments. Second, we focus the use of AI - from light models to deep reinforcement learning - in Character Animation, i.e. towards the training of autonomous 3D characters able to navigate and interact with such environments. The lab sessions are held on Unity, based on C#.
@@ Line 87: / Line 87: @@
 **INF657G - Navigation for Autonomous systems (24h, 2 ECTS), David Filliat (ENSTA)** (contact: david.filliat@ensta-paris.fr)
-   Drones and robots must create maps of their surroundings to plan their movement and navigate. This course presents the robotic platforms and the most common sensors (vision, Lidar, intertial units, odometry …) and the different components of navigation: control; obstacle avoidance; localization; mapping (SLAM) and trajectory planning as well as filtering (Kalman filter, particle filtering, etc.) and optimization techniques used in these fields.
+> Drones and robots must create maps of their surroundings to plan their movement and navigate. This course presents the robotic platforms and the most common sensors (vision, Lidar, intertial units, odometry …) and the different components of navigation: control; obstacle avoidance; localization; mapping (SLAM) and trajectory planning as well as filtering (Kalman filter, particle filtering, etc.) and optimization techniques used in these fields.
 **INF641 - Introduction to the verification of neural networks (24h, 2 ECTS), Eric Goubault (EP), Sylvie Putot (EP)** (contact: sylvie.putot@polytechnique.edu)
-   Neural networks are widely used in numerous applications including safety-critical ones such as control and planning for autonomous systems. A central question is how to verify that they are correct with respect to some specification. Beyond correctness or robustness, we are also interested in questions such as explainability and fairness, that can in turn be specified as formal verification problems. In this course, we will see how formal methods approaches introduced in the context of program verification can be leveraged to address the verification of neural networks.
+> Neural networks are widely used in numerous applications including safety-critical ones such as control and planning for autonomous systems. A central question is how to verify that they are correct with respect to some specification. Beyond correctness or robustness, we are also interested in questions such as explainability and fairness, that can in turn be specified as formal verification problems. In this course, we will see how formal methods approaches introduced in the context of program verification can be leveraged to address the verification of neural networks.
 **INF642 - Socio-emotional embodied conversational agents (24h, 2 ECTS), Catherine Pelachaud (CNRS - ISIR), Chloé Clavel (Inria Paris) ** (contact: catherine.pelachaud@sorbonne-universite.fr; chloe.clavel@inria.fr)
-  Many interactive systems, from virtual companions to online retailing, rely on embodied conversational agents. These agents need to reach a good level of communication skills to conduct a conversation with humans and be acceptable and trustworthy by humans. This course will introduce non-verbal behavior models, present models for multimodal dialog, opinion detection and voice quality, explain how to model the agent's emotions and their evolution over time, and present methods for enhancing naturalism with expressive gaze and gestures, realistic animation.
+> Many interactive systems, from virtual companions to online retailing, rely on embodied conversational agents. These agents need to reach a good level of communication skills to conduct a conversation with humans and be acceptable and trustworthy by humans. This course will introduce non-verbal behavior models, present models for multimodal dialog, opinion detection and voice quality, explain how to model the agent's emotions and their evolution over time, and present methods for enhancing naturalism with expressive gaze and gestures, realistic animation.
 **INF644 - Virtual/Augmented Reality & 3D Interactions  (24h, 2 ECTS), Anatole Lécuyer (Inria Rennes), Ferran Argelaguet (Inria Rennes), Arnaud Prouzeau (Inria Saclay), Claudio Pacchierotti (CNRS - IRISA), Fabien Lotte (Inria Bordeaux)** (contact: anatole.lecuyer@inria.fr)
-  Metaverse and virtual/augmented reality technologies are spreading widely. But reconstructing our world and generating virtual ones would be useless without effective techniques to navigate and interact with them. This course will present virtual and augmented reality systems as well as the associated methods for 3D interaction, from multi-modal interaction merging visual immersion, sound and haptics systems to brain-computer interfaces.
+> Metaverse and virtual/augmented reality technologies are spreading widely. But reconstructing our world and generating virtual ones would be useless without effective techniques to navigate and interact with them. This course will present virtual and augmented reality systems as well as the associated methods for 3D interaction, from multi-modal interaction merging visual immersion, sound and haptics systems to brain-computer interfaces.
 **INF634 - Computer Vision  (24h, 2 ECTS), Vicky Kalogeiton (EP)** (contact: vicky.kalogeiton@polytechnique.edu)
-  This course is an introduction to fundamental and advanced topics in computer vision with learning-based approaches, ie. Deep Learning. Topics include image and video classification, object detection, action recognition, optical flow and motion, multi-modal vision systems, annotation signal and applications.
+> This course is an introduction to fundamental and advanced topics in computer vision with learning-based approaches, ie. Deep Learning. Topics include image and video classification, object detection, action recognition, optical flow and motion, multi-modal vision systems, annotation signal and applications.
-**INF643 - Soft robots: simulation, fabrication, and control (24h, 2 ECTS),  Christian Duriez (Inria Lille), Sylvain Lefebvre (Inria Nancy) ** (contact: christian.duriez@inria.fr)
+**INF643 - Soft robots: Design, Modeling, Simulation and Control (24h, 2 ECTS),  Christian Duriez (Inria Lille)** (contact: christian.duriez@inria.fr)
-  Soft robotics is a promising novel field, bringing more robustness in robots design and for all tasks involving close interactions with humans, from help to disable people to medical robot. This course will give an introduction to recent advances in soft robotics, including topological optimization for additive fabrication, modeling and control techniques for robots, and will present recent applications in medicine, industry and art.
+> Soft robotics is a promising novel field, bringing more robustness in robots design and for all tasks involving close interactions with humans, from help to disable people to medical robot. This course will give an introduction to recent advances in soft robotics, including design, modeling, simulation and control techniques for robots, and will present recent applications in medicine, industry and art.