-
Happy to share that my
@facebookai internship work "Depth-adaptive Transformer" has been accepted to#ICLR2020 . TL;DR: We dynamically adjust the computation per input and match the accuracy of a baseline Transformer with only 1/4 the decoder layers. https://openreview.net/forum?id=SJg7KhVKPH …Prikaži ovu nit -
Our paper studying the emergent cross-lingual properties of multilingual BERT is accepted to
#ICLR2020! Lots of dedicated work from (undergrads) Karthikeyan K and Zihan Wang. tl;dr it's network depth, not wordpiece overlap. https://openreview.net/forum?id=HJeT3yrtDr … -
Really exciting to have my first paper accepted at
@iclr_conf#ICLR2020! It provides the first group theoretical approach towards equivariant visual attention. Nice things coming up next!@erikjbekkers@jmtomczak@Mark. Co-Attentive Equivariant Nets: https://openreview.net/forum?id=r1g6ogrtDr …pic.twitter.com/jppwAD8NJy
-
#ICLR2020 authors with > 5 submissions: 32 Sergey Levine 20 Yoshua Bengio 16 Cho-jui Hsieh 14 Pieter Abbeel 13 Liwei Wang, Tom Goldstein, Chelsea Finn, Bo Li, Jun Zhu # of accepted papers: 13 Sergey Levine 7 Le Song, Jun Zhu 6 Cho-jui Hsieh, Jimmy Ba, Liwei Wang, Pushmeet Kohli pic.twitter.com/FzIfcojhhx
-
Finally ...... our paper on "foresight pruning" just got accepted by
@iclr_conf. We introduced a simple, yet effective pruning criterion for pruning networks before training and related the criterion to recent NTK analysis.#ICLR2020 https://openreview.net/forum?id=SkgsACVKPH …Prikaži ovu nit -
Our paper (joint work with my supervisor
@btreetaiji ) has been accepted as a spotlight (48 long talks and 108 spotlights out of 2594 submissions) at#ICLR2020 !! https://openreview.net/forum?id=S1ldO2EFPr … -
We're pleased to let you know that your submission, Exploration in Reinforcement Learning with Deep Covering Options, has been accepted at
#ICLR2020! https://openreview.net/forum?id=SkeIyaVtwB … This work was led by@DINDIN92, with Jee Won Park and George Konidaris. More
Prikaži ovu nit -
Our paper ‘Progressive Learning and Disentanglement of Hierarchical Representations’ spearheaded by Zhiyuan has been accepted at
#ICLR2020 (oral). TL;DR: strategy to progressively learn disentangled hierarchical representations+new disentanglement metric! https://openreview.net/forum?id=SJxpsxrYPS … -
peer review in machine learning is broken
#ICLR2020 -
In ICLR last year, we show that existing few-shot classification methods perform poorly across different domains (https://bit.ly/34GzQrl ). This year in
#ICLR2020, we show how we can make few-shot learners generalize better to unseen domains (https://bit.ly/35HJATL )! -
Excited to share that our paper "Semi-Supervised Generative Modeling for Controllable Speech Synthesis" got accepted at
#ICLR2020#ICLR paper: https://openreview.net/forum?id=rJeqeCEtvH … demo: https://google.github.io/tacotron/publications/semisupervised_generative_modeling_for_controllable_speech_synthesis/index.html …Prikaži ovu nit -
Accepted to
@iclr_conf#ICLR2020: Settling permutation equivariance universality for popular deep models including DeepSets and some versions of PointNet.@SegolNimrodhttps://openreview.net/forum?id=HkxTwkrKDB … -
Our submission to
#ICLR2020 was accepted as a spotlight, in it we try to clear up some confusion about Bayesian inference in RL, read it if you want to understand how Bayes and Bellman can get along! https://twitter.com/IanOsband/status/1208394091303784450 …
-
Decisions released
Congratulations to accepted papers; to those who we could not accommodate, we wish you success in your ongoing research. See our blog for the first of our reflections. See you soon in Ethiopia. #ICLR2020#OurHatata
https://link.medium.com/ysHgHxoLy2 -
Mathematical Reasoning in Latent Space https://openreview.net/forum?id=Ske31kBtPr … will be featured at
#iclr2020. Multiple step reasoning can be performed on embedding vectors of mathematical formulas. -
1/ New paper on an old topic: turns out, FGSM works as well as PGD for adversarial training!* *Just avoid catastrophic overfitting, as seen in picture Paper: https://arxiv.org/abs/2001.03994 Code: https://github.com/locuslab/fast_adversarial … Joint work with
@_leslierice and@zicokolter to be at#ICLR2020 pic.twitter.com/2EmwFaX7Qp
Prikaži ovu nit -
How much supervision do you need to learn disentangled representations? Turns out, not that much! Joint work with
@mtschannen, S. Bauer,@gxr,@bschoelkopf and@OlivierBachem. Accepted at#ICLR2020@GoogleAI@ETH_en@MPI_IShttps://openreview.net/forum?id=SygagpEKwB … -
Our
#ICLR2020 paper "What can neural networks reason about?" in 3 slides, presented at Winter Festa Episode 5 on Christmas. Great discussion with researchers from all over Japan! See Stefanie's talks for more detail https://bit.ly/2t7n2Nu (NeurIPS), https://bit.ly/2PYBd0u (IAS)pic.twitter.com/Zc2Rb5nl4E
-
#ICLR2020 announcement time!@mark_riedl and I first use KGs for deep POMDP agents https://arxiv.org/abs/1812.01628 Then KGs for commonsense transfer https://arxiv.org/abs/1908.06556 Now, KGs+RL for combinatorially sized language action spaces w/@mhauskn@MSFTResearchhttps://openreview.net/forum?id=B1x6w0EtwH …Prikaži ovu nit -
Really pleased that our paper "Multiplicative Interactions and Where to Find Them" was accepted to
#ICLR2020. https://openreview.net/forum?id=rylnK6VtDH … We present a unifying picture of several different neural net architectural motifs, that all involve multiplicative interactions... 1/3pic.twitter.com/G8qetMMidx
Prikaži ovu nit
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.