Learn More
We present an integrated framework for using Convolutional Networks for classification , localization and detection. We show how a multiscale and sliding window approach can be efficiently implemented within a ConvNet. We also introduce a novel deep learning approach to localization by learning to predict object boundaries. Bounding boxes are then(More)
This article offers an empirical exploration on the use of character-level convolu-tional networks (ConvNets) for text classification. We constructed several large-scale datasets to show that character-level convolutional networks could achieve state-of-the-art or competitive results. Comparisons are offered against traditional models such as bag of words,(More)
" Physics of Optoelectronic and This dissertation explores the fundamental interactions between light and matter towards devices applications in the field of Opto-electronics and metal-optics, or plasmonics. In its core, this dissertation attempts, and succeeds to demonstrate strong enhancements of such interactions. Here, surface plasmon polaritons,(More)
Integrated optical modulators with high modulation speed, small footprint and large optical bandwidth are poised to be the enabling devices for on-chip optical interconnects. Semiconductor modulators have therefore been heavily researched over the past few years. However, the device footprint of silicon-based modulators is of the order of millimetres, owing(More)
How organ-specific metastatic traits arise in primary tumors remains unknown. Here, we show a role of the breast tumor stroma in selecting cancer cells that are primed for metastasis in bone. Cancer-associated fibroblasts (CAFs) in triple-negative (TN) breast tumors skew heterogeneous cancer cell populations toward a predominance of clones that thrive on(More)
This article demonstrates that we can apply deep learning to text understanding from character-level inputs all the way up to abstract text concepts, using temporal convolutional net-works(LeCun et al., 1998) (ConvNets). We apply ConvNets to various large-scale datasets, including ontology classification, sentiment analysis, and text categorization. We show(More)
As a promising tool for identifying genetic markers underlying phenotypic differences, genome-wide association study (GWAS) has been extensively investigated in recent years. In GWAS, detecting epistasis (or gene-gene interaction) is preferable over single locus study since many diseases are known to be complex traits. A brute force search is infeasible for(More)
Transfer learning aims to leverage the knowledge in the source domain to facilitate the learning tasks in the target domain. It has attracted extensive research interests recently due to its effectiveness in a wide range of applications. The general idea of the existing methods is to utilize the common latent structure shared across domains as the bridge(More)
A long-term goal of machine learning is to build intelligent conversational agents. One recent popular approach is to train end-to-end models on a large amount of real dialog transcripts between humans (Sordoni et al., 2015; Vinyals & Le, 2015; Shang et al., 2015). However, this approach leaves many questions unanswered as an understanding of the precise(More)