|
Alec Tschantz
@
a_tschantz
Brighton, England
|
|
PhD student with the Sackler Center for Consciousness Science.
|
|
|
476
Tweetovi
|
1.549
Pratim
|
439
Osobe koje vas prate
|
| Tweetovi |
| Alec Tschantz proslijedio/la je tweet | ||
|
Julian Togelius
@togelius
|
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?
arxiv.org/abs/2001.09908 pic.twitter.com/7bCtBp8xUG
|
||
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Andrei Bursuc
@abursuc
|
27. sij |
|
TRADI: Tracking deep neural network weight distributions -- work with G. Franchi arxiv.org/abs/1912.11316 We’re proposing a cheap method for getting ensembles of networks from a single network training 1/ pic.twitter.com/J31E9aaiKL
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
IvilinStoianov
@IvilinStoianov
|
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" biorxiv.org/content/10.110…
|
||
|
|
||
|
|
Alec Tschantz
@a_tschantz
|
20. sij |
|
I don't think the work says anything about Markov blankets per se, rather that the existence of a Markov blanket does not necessarily imply a Bayesian interpretation of internal dynamics
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Quantitative Biology
@BioPapers
|
20. sij |
|
A technical critique of the free energy principle as presented in "Life as we know it". arxiv.org/abs/2001.06408
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
DeepMind
@DeepMind
|
15. sij |
|
We worked with @harvardbrainsci 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: deepmind.com/blog/article/D… (2/2)
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Warrick Roseboom
@RoseboomWarrick
|
15. sij |
|
ICYMI: new preprint by @maxine_sherman @zfountas @anilkseth 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 @TimingForum biorxiv.org/content/10.110…
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
hardmaru
@hardmaru
|
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.”
cims.nyu.edu/~andrewgw/case… pic.twitter.com/86rV2eqqXD
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Jaan Aru
@jaaanaru
|
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 @sciencemagazine. Congrats Albert, Tim @mattlark @YiotaPoirazi & CO science.sciencemag.org/content/367/64…
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Stefan Frässle
@stefan_fraessle
|
28. pro |
|
"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 royalsocietypublishing.org/doi/full/10.10…
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Marlos C. Machado
@MarlosCMachado
|
20. pro |
|
We're pleased to let you know that your submission, On Bonus Based Exploration Methods In The Arcade Learning Environment, has been accepted at #ICLR2020!
openreview.net/forum?id=BJewl…
This huge endeavor was led by @aalitaiga. W/ @LiamFedus, @marcgbellemare & @AaronCourville.
More👇🏼
|
||
|
|
||
|
|
Alec Tschantz
@a_tschantz
|
19. pro |
|
I think I've misunderstood parts of this conversation - but yes, you can maximise model evidence without changing data
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
|
DeepMind
@DeepMind
|
16. pro |
|
DeepInsight is a decoding framework for discovering and characterising the neural correlates of behaviour and stimuli in unprocessed biological data:
biorxiv.org/content/10.110… pic.twitter.com/K6goibHlha
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Sergey Levine
@svlevine
|
30. lis |
|
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, @chelseabfinn, J. Wu, J. Tenenbaum
sites.google.com/view/op3websit… pic.twitter.com/9OJ3fgr1OS
|
||
|
|
||
|
|
Alec Tschantz
@a_tschantz
|
8. pro |
|
Humans maybe, but what about a network with two hidden nodes? If that doesn’t fit the criteria but a network with N nodes does, what’s the magic sauce?
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Vicarious
@vicariousai
|
7. pro |
|
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. openreview.net/pdf?id=rJeoKJ3… pic.twitter.com/SVR0kxpwQJ
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Dileep George
@dileeplearning
|
7. pro |
|
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? #tweeprint...(1) @vicariousai @swaroopgj @rvrikhye biorxiv.org/content/10.110… twitter.com/vicariousai/st…
|
||
|
|
||
|
|
Alec Tschantz
@a_tschantz
|
7. pro |
|
haha no way I didn’t even notice the name! hope you are well
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
Danijar Hafner
@danijarh
|
4. pro |
|
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 arxiv.org/pdf/1912.01603…
code github.com/google-researc…
Thread 👇 pic.twitter.com/K5DnooVIUH
|
||
|
|
||
| Alec Tschantz proslijedio/la je tweet | ||
|
OpenAI
@OpenAI
|
3. pro |
|
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: openai.com/blog/procgen-b… pic.twitter.com/OhECCCAeY3
|
||
|
|
||