Skip to search formSkip to main contentSkip to account menu
  • Publications
  • Influence
Multi-resolution Networks For Flexible Irregular Time Series Modeling (Multi-FIT)
TLDR
A unified model named Multi-resolution Flexible Irregular Time series Network (Multi-FIT) is proposed, which improves upon the state-of-the-art models for three predictive tasks, including the forecasting of patient survival.
View PDF on arXiv
Long Document Summarization in a Low Resource Setting using Pretrained Language Models
TLDR
This paper attempts to compress these long documents by identifying salient sentences in the source which best ground the summary, using a novel algorithm based on GPT-2 language model perplexity scores, that operates within the low resource regime.
View on ACL
Using error decay prediction to overcome practical issues of deep active learning for named entity recognition
TLDR
A transparent batch active sampling framework by estimating the error decay curves of multiple feature-defined subsets of the data that can make the sampling process more robust to labeling noise when combined with uncertainty-based methods is proposed.
View on Springer
Overcoming Practical Issues of Deep Active Learning and its Applications on Named Entity Recognition
TLDR
This work proposes a transparent batch active sampling framework by estimating the error decay curves of multiple feature-defined subsets of the data and demonstrates that the proposed methods significantly outperform diversification-based methods for black-box NER taggers and can make the sampling process more robust to labeling noise when combined with uncertainty- based methods.
View PDF on arXiv
Methods for Various Illumination Settings using Infrared Thermographic Images
TLDR
A system that shall det ect and track vehicles in real time, using near inf rared taken from an infrared thermographic camera, using regular cameras with visible light images as input is described.
Multi-resolution Attention with Signal Splitting for Multivariate Time Series Classification
TLDR
This work presents a unified method which handles all three significant challenges of multivariate time series: Multi-resolution Attention with Signal Splitting (MASS), which can be applied to any existing model, significantly boosting predictive performance.
Leveraging One's Impact Through Digital Persona and Image Management for Enhanced Effectiveness and Competitive Advantage
TLDR
Digital PRIM (D-PRIM) is evolving in a significant way to cater to image enhancement of individuals in the digital era which also impacts the image and success of their organizations.
View on SSRN
LRS-DAG: Low Resource Supervised Domain Adaptation with Generalization Across Domains
TLDR
It is shown that, in the case of FCNs, when transferring from MNIST to SVHN, LRS-DAG performs comparably to fine tuning, with the advantage of maintaining performance over the source domain.
View PDF on arXiv
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy, Terms of Service, and Dataset License