Home /
Research
Showing 229 - 234 / 897
Deeper or Wider Networks of Point Clouds with Self-attention?
HaoxiRan, LiLu....
Published date-11/29/2020
Prevalence of deeper networks driven by self-attention is in stark contrast to underexplored point-based methods. In this paper, we propose groupwise self-attention as the basic block to construct our network: …
Latent Template Induction with Gumbel-CRFs
YaoFu, ChuanqiTan, BinBi....
Published date-11/29/2020
Data-to-TextGeneration, ParaphraseGeneration, TextGeneration
Learning to control the structure of sentences is a challenging problem in text generation. Existing work either relies on simple deterministic approaches or RL-based hard structures. We explore the use …
Scaling down Deep Learning
SamGreydanus....
Published date-11/29/2020
Though deep learning models have taken on commercial and political relevance, many aspects of their training and operation remain poorly understood. This has sparked interest in "science of deep learning" …
Differences between human and machine perception in medical diagnosis
TaroMakino, StanislawJastrzebski, WitoldOleszkiewicz....
Published date-11/28/2020
BreastCancerDetection, MedicalDiagnosis
Deep neural networks (DNNs) show promise in image-based medical diagnosis, but cannot be fully trusted since their performance can be severely degraded by dataset shifts to which human perception remains …
Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules
JohannesKlicpera, ShankariGiri, JohannesT.Margraf....
Published date-11/28/2020
Many important tasks in chemistry revolve around molecules during reactions. This requires predictions far from the equilibrium, while most recent work in machine learning for molecules has been focused on …
Understanding How BERT Learns to Identify Edits
SamuelStevens, YuSu....
Published date-11/28/2020
Pre-trained transformer language models such as BERT are ubiquitous in NLP research, leading to work on understanding how and why these models work. Attention mechanisms have been proposed as a …