Alec Tschantz

@a_tschantz

PhD student with the Sackler Center for Consciousness Science.

Vrijeme pridruživanja: veljača 2018.

Tweetovi

Blokirali ste korisnika/cu @a_tschantz

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

  1. Prikvačeni tweet
    26. stu 2019.

    A new preprint from myself, , and : "scaling active inference":

    Poništi
  2. proslijedio/la je Tweet
    1. velj

    To the left, you see a trained agent playing a level of a game. To the right, you see the same playthrough from an agent-centric perspective: cropped, translated, and rotated with the agent in the center. Which perspective is the best input for the agent?

    Prikaži ovu nit
    Poništi
  3. proslijedio/la je Tweet
    27. sij

    TRADI: Tracking deep neural network weight distributions -- work with G. Franchi We’re proposing a cheap method for getting ensembles of networks from a single network training 1/

    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    21. sij

    Finally it's out: The simplicity of mixture models enhanced with dynamics brings to an elegant and powerful generative computational model. "The hippocampal formation as a hierarchical generative model supporting generative replay and continual learning"

    Poništi
  5. proslijedio/la je Tweet
    20. sij

    A technical critique of the free energy principle as presented in "Life as we know it".

    Poništi
  6. proslijedio/la je Tweet
    15. sij

    We worked with to show that distributional RL, a recent development in AI research, can provide insight into previously unexplained elements of dopamine-based learning in the brain. Read the blog: (2/2)

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    15. sij

    ICYMI: new preprint by and me. In a preregistered model-based fMRI analysis we show that individual human duration estimates for naturalistic videos can be reproduced from visual cortex BOLD

    Prikaži ovu nit
    Poništi
  8. proslijedio/la je Tweet
    12. sij

    The Case for Bayesian Deep Learning ”Bayesian or not, the prior will certainly be imperfect. Avoiding an important part of the modelling process because one has to make assumptions, however, will often be a worse alternative than an imperfect assumption.”

    Poništi
  9. proslijedio/la je Tweet
    2. sij

    Brains are amazing. Our lab demonstrates that single human layer 2/3 neurons can compute the XOR operation. Never seen before in any neuron in any other species. Out now in . Congrats Albert, Tim  & CO

    Poništi
  10. proslijedio/la je Tweet
    28. pro 2019.

    "Markov blankets, information geometry and stochastic thermodynamics" - new work from Thomas Parr, Lancelot da Costa and Karl Friston demonstrating the link between thermodynamics, information and inference

    Poništi
  11. proslijedio/la je Tweet
    20. pro 2019.

    We're pleased to let you know that your submission, On Bonus Based Exploration Methods In The Arcade Learning Environment, has been accepted at ! This huge endeavor was led by . W/ , & . More👇🏼

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    16. pro 2019.

    DeepInsight is a decoding framework for discovering and characterising the neural correlates of behaviour and stimuli in unprocessed biological data:

    Prikaži ovu nit
    Poništi
  13. proslijedio/la je Tweet
    29. lis 2019.

    Model-based RL with models that factorize over entities; can discover object-like representations, and can be used to plan how to construct structures out of parts. w/ R. Veerapaneni, JD Co-Reyes, M. Chang, M. Janner, , J. Wu, J. Tenenbaum

    Poništi
  14. proslijedio/la je Tweet
    6. pro 2019.

    You'd think a generative model should answer different queries but deep learning models like VAEs are trained to answer just one type of query! We introduce a method called query-training that creates an inference network that can answer novel query types.

    Poništi
  15. proslijedio/la je Tweet
    6. pro 2019.

    Are you skeptical about successor representations? Want to know how our new model can learn cognitive maps, context-specific representations, do transitive inference, and flexible hierarchical planning? ...(1)

    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    4. pro 2019.

    We introduce Dreamer, an RL agent that solves long-horizon tasks from images purely by latent imagination inside a world model. Dreamer improves over existing methods across 20 tasks. paper code Thread 👇

    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    3. pro 2019.

    We're releasing Procgen Benchmark, 16 procedurally-generated environments for measuring how quickly a reinforcement learning agent learns generalizable skills. This has become the standard research platform used by the OpenAI RL team:

    Poništi
  18. proslijedio/la je Tweet
    3. pro 2019.

    What's Hidden in a Randomly Weighted Neural Network? “Hidden in a randomly weighted Wide ResNet-50 we show that there is a subnetwork (with random weights) that is smaller than, but matches the performance of a ResNet-34 trained on ImageNet.” 😮

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    1. pro 2019.

    I have been waiting for this paper by and Tononi. Looking forward to reading it in detail. Why Does Space Feel the Way it Does? Towards a Principled Account of Spatial Experience

    Poništi
  20. proslijedio/la je Tweet
    2. pro 2019.

    The interaction between machine learning & neuroscience isn't just limited to sensory perception or cognition! New review on "Hierarchical motor control in mammals and machines"

    Poništi
  21. proslijedio/la je Tweet
    2. pro 2019.

    Structural knowledge allows participants to infer unobserved transitions and beneficial policy. Basis sets for structural knowledge allow these inferences by generalising over task particularities. by me Thomas Parr Steve Kennerley

    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:

    ·