Skip to content
View oublalkhalid's full-sized avatar
🎯
Focusing
🎯
Focusing
35 followers · 127 following

Achievements

Achievement: YOLOAchievement: StarstruckAchievement: Pull SharkAchievement: Quickdraw

Achievements

Achievement: YOLOAchievement: StarstruckAchievement: Pull SharkAchievement: Quickdraw

Block or report oublalkhalid

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
oublalkhalid/README.md

Hi! I'm Khalid

  • 👀 I am interested in representation learning, causality, and contextual bandits, with a focus on the theoretical foundations of reliable and interpretable deep learning with large models. My research is motivated by fundamental questions about generalization, robustness, and scientific understanding, with applications to weather forecasting, sustainability, and Earth system modeling. My thesis, “From Signal to Structure,” develops this direction through identifiability theory, causality, invariance, and equivariance, studying when meaningful representations can be theoretically recovered from observational data and used for decision-making under uncertainty

  • 🥇 MS.E in Computer Science (summa cum laude) from Ecole Polytechnique - Institut Polytechnique of Paris, France. Currently, I follow a Ph.D. program at the same institute.

  • 💞️ Open to collaborate on Explainability for Generative Models

  • 📚 What I have read ?

  • 📫 How to reach me khlaid.oublal@polytechnique.edu (.org [for graduate email]) | or khlaid.oublal@ip-paris.fr

  • Training at Mathematical Institute, University of Oxford

  • Summer School Oxford, Machine Learning (OxML2023): Generative Models

  • Current work:

    • Satisfiability modulo theories, Neural networks as a sub-symbolic approach with Pr. Sergio Mover
    • Deep Q-Learning systems to avoid collisions 802.11bf electric scooter with Pr. Keun-Woo Lim
    • Explainable Models for sequential data with Pr. François Roueff and Pr. Said Ladjal. Follow-up by Pr. Cristian Jutten.
    • OpenXAI for time series with Stanford University (ongoing...)

  • I collaborate to @huggingface Time Series Transformer

News 📣:

  • Working on Forecasting of weather data and solving inverse problems using Generative Models
  • [January 2024]🚀 Paper accepted at ICLR 2024: Disentangling Time Series Representations via Contrastive Independence-of-Support on l-Variational Inference
  • [December 2023] Paper Spotlight in https://neurips.cc/virtual/2023/83222
  • [September 2023] Paper accepted at NeurIPS 2023: DISCOV
  • [March 2023] Paper accepter at ICML 2023: Temporal Attention Bottleneck is Informative?

Skills

angularjs aws bash csharp cypress django docker dotnet elasticsearch express flask git go java javascript jenkins kafka kubernetes linux mongodb mysql nodejs php postgresql postman python rabbitMQ react redis travisci typescript zapier

Pinned Loading

  1. TimeSAE TimeSAE Public

    [Accepted at ICML 2026 🎉 ] TimeSAE: Sparse Decoding for Faithful Explanations of Black-Box Time Series Models

    Python 3

  2. MStar-Diffusion MStar-Diffusion Public

    [Accepted at ICML 2026 🎉 ] Markovian Projection of Star-Shaped Diffusion for Exponential Family Distributions

    Python 1

  3. Institut-Polytechnique-de-Paris/time-disentanglement-lib Institut-Polytechnique-de-Paris/time-disentanglement-lib Public template

    🤗 [ICLR 2024] Disentangling Time Series Representations via Contrastive based l-Variational Inference

    Python 20 2

  4. NVIDIA/physicsnemo NVIDIA/physicsnemo Public

    Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods

    Python 3k 724

  5. MoroccoAI-Data-Challenge-Nvidia-ANRT-MoroccoAI MoroccoAI-Data-Challenge-Nvidia-ANRT-MoroccoAI Public

    MoroccoAI Challenge - This paper won the Nvidia - ANRT - MoroccoAI (conference of December 2021)

    Jupyter Notebook 14 1

  6. huggingface/transformers huggingface/transformers Public

    🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

    Python 163k 33.9k