Robert Lange

@RobertTLange

Deep RL PhD Student@TU Berlin. Intelligence. Biological&Artificial {🐠 ♥️ 🦈}

Vrijeme pridruživanja: travanj 2017.

Tweetovi

Blokirali ste korisnika/cu @RobertTLange

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @RobertTLange

  1. Prikvačeni tweet
    15. sij

    🎉 2019 🎉 was quite the year for Deep Reinforcement Learning. In todays blog post I list my top 10 papers 🦄💻🧠 What was your favourite paper? Let me know!

    Poništi
  2. 29. sij

    Another day at the Lab: “Blitzforschung“ = 5min blackboard presentation of a paper 🤓. Yesterday‘s gem: Backpropamine by , et al. (2019; ICLR). Dopamine-inspired modulation of differential plasticity = Bridging 🌉 timescales in Meta-Learning 🧠

    Poništi
  3. 22. sij

    🎉 Celebrating 10 'flexible learning' reading group sessions ! What better way to do so, than with some deep neuroethology 🧠🐀🤖? Checkout all presentations & join the mailing list 👉

    Poništi
  4. 20. sij

    Interested in principled off-policy correction? outlines how to combat covariate shift in RL. Work (P3O & Meta-Q-Learning) done with and presented 🧠🦾✌️

    Poništi
  5. 10. sij

    Two days ago gave a great presentation on “Recurrent Independent Mechanisms” by , , et al. 19’ at the Flexible Learning Reading Group . Checkout the slides & join the mailing list🧠🙈✌️:

    Poništi
  6. 8. sij

    Great kick-off to the new year: presents his fundamental work on Multi-Agent Learning from DIAL & Differentiable Communication Protocols to Multi-Agent Search in Hanabi ! 🦾🧠🧑‍💻

    Poništi
  7. 15. pro 2019.

    What a week 🧠🤓💻! I loved meeting so many of you at - the ML community is truly wonderful. Checkout all my collected visual notes ✍️ & feel free to share:

    Poništi
  8. 15. pro 2019.

    Wow! This is amazing. Make sure to check out extensive conference notes from . Thank you for sharing! 🙏✍️

    Poništi
  9. 14. pro 2019.

    ‘Innate everything‘ 🧠🧐🐊 - argues for the importance of finding the right inductive biases in bodies/architectures (WANNs) & prediction/world models (Observational Dropout) - Transferable Skills Workshop

    Poništi
  10. 14. pro 2019.

    And... another one. Curious about how to combine Hierarchical RL and Grammar Induction to define a powerful Action Grammar? Checkout my work with at the Deep RL & Transferable Skills Workshops. Preprint:

    Poništi
  11. 14. pro 2019.

    Great presentation of the by , , & the PPO-LSTM-based winning solution by . Very much looking forward to the next generation including human & animal expert traces!

    Poništi
  12. 14. pro 2019.

    Many gems in Deep RL workshop talk at on AlphaStar. Including scatter connections, imitation-based regularization, the league & the unique problem decomposition.

    Poništi
  13. 13. pro 2019.

    Workshop talks should push conceptual limits. Fascinating talk by Rich Sutton at the Bio&Artificial RL workshop P.S.: I will do my best 🧠🧐✍️

    Poništi
  14. 13. pro 2019.

    Tackling optimization (WarpGrad by ) & memory (GrTrXL by E. Parisotto) in meta-learning - a summary of ’s meta-learning workshop talk at .

    Poništi
  15. 13. pro 2019.

    Towards useful abstractions in RL - great talk on state & action abstractions that trade-off compression and value preservation by at the meta-learning workshop .

    Poništi
  16. 13. pro 2019.

    Part I of Bio&Artificial RL Workshop at is in the books. Great set of talks covering different timescales of learning/structured priors through meta-learning (), intrinsic motivation (J. Gottlieb) as well as planning through cognitive maps ().

    Poništi
  17. 12. pro 2019.

    Guess who is the new lucky owner of an awesome art piece by ? It was great to get to know more about the ML-assisted generation process. Check out his projects & ‚ML for Creativity & Design’ workshop at .

    Poništi
  18. 12. pro 2019.

    Become part of 2020 in Poland - and next years meet-up! Checkout this blog post covering my experience & feel free to get in touch:

    Poništi
  19. 11. pro 2019.

    Yoshua Bengio - „It is Time for ML to explore Consciousness“ 🧠👌🧐

    Poništi
  20. 11. pro 2019.

    Truly eye-opening keynote by at . Intelligence does not only evaluate functions. Towards a biologically-plausible formulation of social intelligence 🤖🧠🕵️‍♀️

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·