cv
Basics
| Name | Takieddine Soualhi |
| Label | Research Engineer in AI & Robotics |
| takieddine.soualhi@inria.fr | |
| Url | https://takieddinesoualhi.github.io |
| Summary | Postdoctoral research engineer in AI and Robotics at Inria Lyon (CHROMA team), working on multi-agent reinforcement learning for human-robot coordination. |
Work
-
2025.04 - Present Research Engineer
INRIA, CHROMA Team
Introducing novel MARL methods for human-agent coordination.
- Introducing novel MARL methods for human-agent coordination.
- Conducting large scale simulations and deploying trained models for industrial partner (Enchanted Tools).
- Integration of the ROS2 navigation stack into the partner's humanoid robot.
-
2022.02 - 2024.02 Teaching Assistant
University of Burgundy and Franche-Comté
Teaching courses in computer science and robotics.
- Courses: C Programming, Operating Systems, Web Applications, Robotics.
- Supervision of students semester projects.
-
2021.11 - 2025.02 Research Engineer / Ph.D. candidate
CIAD lab
PhD research on deep reinforcement learning for mobile robot visuomotor control.
- Conduct literature review, identify literature, and set scientific goals.
- Build custom RL training environments and training robots.
- Propose novel RL and MARL algorithms for training single and multi-robot systems on vision-based control tasks.
- Deploy the trained models on real robots.
- Publish results in top-tier robotics journals.
-
2021.04 - 2021.06 Research Engineer
LoopAI
Research on facial expression recognition.
- Literature review on facial expression recognition.
- Evaluation of CNN and GAN-based models on the FER dataset for facial expression recognition.
-
2021.03 - 2021.09 Machine Learning Engineer
LINEACT lab
Machine learning for human-robot interaction datasets.
- Preprocessing and cleaning the InHard multimodal human-robot interaction dataset.
- Collecting and building a custom dataset using MoCap data in VR.
- Proposing, implementing, and evaluating deep learning models for both the InHARD dataset and the newly collected dataset.
- Shipping code along with documentation.
-
2019.10 - 2020.02 Research Assistant
Sorbonne University / ISIR lab
Research on autonomous driving simulation and deep learning.
- Building a simulation scene and its corresponding real scene for autonomous driving.
- Fine-tuning deep learning models for scene understanding.
- Implementing control and planning algorithms.
- Shipping code along with demo of the platform.
-
2015.11 - 2017.09 Mentor
AEPI NewVision
Mentoring students in DIY robotics and electronics projects.
- Building DIY projects, including robots, electronic boards, IoT templates.
- Mentoring students and supervising DIY projects.
- Presenting projects and participating in conferences.
Education
-
2021.11 - 2024.12 Belfort, France
Ph.D.
University of Burgundy and Franche-Comté
Computer Science
- Dissertation: Deep reinforcement learning based visuomotor policies for mobile robots in single and multi-agent configurations.
- Keywords: Reinforcement learning, Deep learning, Robotics, Computer vision, Multi-agent systems.
- Supervisors: Prof. Yassine Ruichek, Prof. Stéphane Galland.
-
2018.09 - 2021.09 Paris, France
M.Sc.
Sorbonne University
Intelligent Systems and Robotics
- Keywords: Robotics, Audio processing, Computer vision, Machine learning, HCI.
Awards
- 2020.09.01
Paris Cité University Scholarship for International Talents
Paris Cité University
Scholarship for International Talents - Fall 2020.
- 2019.06.01
BioMed-Innov Scholarship for Innovation in the Biomedical Field
BioMed-Innov
Scholarship for Innovation in the Biomedical Field - Summer 2019.
Skills
| Reinforcement Learning | |
| Dynamic programming | |
| Value based RL | |
| Policy gradients | |
| Actor-critic methods | |
| DQN, PPO, SAC, Rainbow, R2D2, Agent57 | |
| Behaviour cloning, DAgger, RLHF, IRL, GAIL |
| Deep Learning | |
| MLPs, CNNs, RNNs, Attention models, Transformers | |
| Dropout, L1 and L2, Data augmentation, Batch normalization | |
| VLMs, LLMs, VLAs (Pi0, SmolVLA, RT-x) |
| Robotics | |
| State Estimation, SLAM | |
| Visual servoing, Path planning, Robot modelling | |
| Gazebo, Pybullet, Habitat-sim, Gibson, Isaac-sim | |
| ROS1/2 | |
| Turtlebot3, Robotnik Summit-XL |
Languages
| Arabic | |
| Native speaker |
| French | |
| Fluent |
| English | |
| Fluent |
Interests
| Research Interests | |
| Reinforcement Learning | |
| Multi-Agent Systems | |
| Robotics | |
| Computer Vision | |
| Human-Robot Interaction |