Adversarial Training Methods For Semi-Supervised Text Classification

Introduction This paper is the first work that applies adversarial training and virtual adversarial training to sequence models, and it greatly improved text classification tasks. In applying the adversarial training, this paper adopts distributed word representation, or word embedding, as the input, rather than the traditional one-hot representation. The reason lies in the fact that […]

Continue reading


2017 Trends — What You Want & What Comes

Source: c’t Magazin für Computer Technik 3/17 [Vedio Source] [Article Source] Don’t be worried about AI. A glance at the state of the art research shows that neural networks would still serve us, and artificial general intelligence is not yet in sight. Therefore, robots and language assistants are nowhere near as smart as what the […]

Continue reading


Yann Le Cun: Predicting under Uncertainty, the Next Frontier in AI

Seminar Info: http://datascience.inf.ed.ac.uk/events/data-science-distinguished-lecture/ Hello everyone. I’m so happy for this opportunity to share the valuable things I learned from professor Yann LeCun’s lecture at the University of Edinburgh on January 13, 2017. Unfortunately, I did not understand everything mentioned in the lecture, but I will try my best to share what I can. Abstract: The […]

Continue reading