Çağatay Yıldız

AI Research Building

Maria-von-Linden-Str. 6

72076 Tuebingen, Germany

I’m a postdoctoral researcher at the Bethge Lab, University of Tuebingen. I received my doctoral degree from the Computational Systems Biology research group, supervised by Harri Lähdesmäki at Aalto University, Finland. Before that, I worked with Taylan Cemgil in my Master’s degree at Bogazici University, Istanbul.

My postdoctoral studies involve different aspects of continual learning. In my PhD, I tried to understand how continuous-time models can be used in time-series learning, video prediction and reinforcement learning. Curious about the connections with optimal control. My research interests broadly include

  • Continual learning
  • Neural ordinary differential equations
  • Generative models
  • Reinforcement learning

My name is super easy to pronounce: chaa-tai.

news

Oct 6, 2022 👨‍💻 Paper submitted to ICLR 2023!
Oct 6, 2022 🥳 Interacting Dynamical Systems paper accepted to NeurIPS 2022!
Jul 23, 2022 🎤 Talk on continuous-time RL at ICML 2022 workshop on continuous-time methods.
Jun 17, 2022 🎤 Talk on neural ODEs at Nordic Probabilistic AI School. Lecture material here.
Jun 1, 2022 👨‍💻 Starting as a postdoc in Bethge Lab in Tuebingen!
May 16, 2022 🎉 Paper accepted to UAI 2022, read it here
Apr 27, 2022 🎤 Talk on neural ODEs at Linkoping University ML seminars.
Feb 18, 2022 🎓 Successfully defended my dissertation! Read it here
Sep 1, 2021 👨‍💻 Visiting Bosch AI, Renningen, Germany to work with Barbara Rakitsch
Jun 18, 2021 🎤 Talk on ODE2VAE at Nordic Probabilistic AI School.

selected publications

  1. submitted
    Latent Neural ODEs with Sparse Bayesian Multiple Shooting
    Iakovlev, Valerii, Yıldız, Çağatay, Heinonen, Markus, and Lähdesmäki, Harri
    In International Conference on Learning Representations 2022
  2. NeurIPS
    Learning Interacting Dynamical Systems with Latent Gaussian Process ODEs
    Yıldız, Çağatay, Kandemir, Melih, and Rakitsch, Barbara
    In Advances in Neural Information Processing Systems 2022
  3. UAI
    Variational multiple shooting for Bayesian ODEs with Gaussian processes
    Hedge, Pashupati, Yıldız, Çağatay, Lahdesmaki, Harri, Kaski, Samuel, and Heinonen, Markus
    In Uncertainty in Artificial Intelligence 2022
  4. thesis
    Differential Equations for Machine Learning
    Yıldız, Çağatay
    2022
  5. ICML
    Continuous-time Model-based Reinforcement Learning
    Yıldız, Çağatay, Heinonen, Markus, and Lahdesmaki, Harri
    In International Conference on Machine Learning 2021
  6. NeurIPS
    ODE2VAE: Deep generative second order ODEs with Bayesian neural networks
    Yıldız, Çağatay, Heinonen, Markus, and Lahdesmaki, Harri
    In Advances in Neural Information Processing Systems 2019
  7. ICML
    Learning unknown ODE models with Gaussian processes
    Heinonen, Markus, Yıldız, Çağatay, Mannerström, Henrik, Intosalmi, Jukka, and Lähdesmäki, Harri
    In International Conference on Machine Learning 2018
  8. ICML
    Asynchronous Stochastic Quasi-Newton MCMC for Non-Convex Optimization
    Simsekli, Umut, Yıldız, Çağatay, Nguyen, Than Huy, Cemgil, Taylan, and Richard, Gael
    In International Conference on Machine Learning 2018