Home /
Research
Showing 385 - 390 / 897
Low-Dimensional Manifolds Support Multiplexed Integrations in Recurrent Neural Networks
ArnaudFanthomme, RémiMonasson....
Published date-11/20/2020
We study the learning dynamics and the representations emerging in Recurrent Neural Networks trained to integrate one or multiple temporal signals. Combining analytical and numerical investigations, we characterize the conditions …
Learning Informative Representations of Biomedical Relations with Latent Variable Models
HarshilShah, JulienFauqueur....
Published date-11/20/2020
LatentVariableModels, RelationExtraction
Extracting biomedical relations from large corpora of scientific documents is a challenging natural language processing task. Existing approaches usually focus on identifying a relation either in a single sentence (mention-level) …
Concentration inequality for U-statistics of order two for uniformly ergodic Markov chains, and applications
QuentinDuchemin, YohanndeCastro, ClaireLacour....
Published date-11/20/2020
We prove a new concentration inequality for U-statistics of order two for uniformly ergodic Markov chains. Working with bounded $\pi$-canonical kernels, we show that we can recover the convergence rate …
Action Duration Prediction for Segment-Level Alignment of Weakly-Labeled Videos
RezaGhoddoosian, SaifSayed, VassilisAthitsos....
Published date-11/20/2020
This paper focuses on weakly-supervised action alignment, where only the ordered sequence of video-level actions is available for training. We propose a novel Duration Network, which captures a short temporal …
Born Identity Network: Multi-way Counterfactual Map Generation to Explain a Classifier's Decision
KwanseokOh, JeeSeokYoon, Heung-IlSuk....
Published date-11/20/2020
There exists an apparent negative correlation between performance and interpretability of deep learning models. In an effort to reduce this negative correlation, we propose Born Identity Network (BIN), which is …
Graph Signal Recovery Using Restricted Boltzmann Machines
AnkithMohan, AiichiroNakano, EmilioFerrara....
Published date-11/20/2020
Denoising
We propose a model-agnostic pipeline to recover graph signals from an expert system by exploiting the content addressable memory property of restricted Boltzmann machine and the representational ability of a …