What’s New in Deep Learning Research: Inside Google’s Semantic Experiences

What’s New in Deep Learning Research: Inside Google’s Semantic Experiences

  • May 14, 2018
Table of Contents

What’s New in Deep Learning Research: Inside Google’s Semantic Experiences

Last week Google Research made news with the release of Semantic Experiences, a website that serves as a playground to evaluate some of the new advancements in natural language understanding(NLU) technologies. The initial release included two pseudo-games that illustrates the practical viability of some of Google’s latest NLU research. The second initiatives included in Semantic Experiences is Semantris, a game powered by machine learning, where you type out words associated with a given prompt.

The experience might seem trivial but, speaking as an end user, I can tell you it can become addictive. Google’s Talk to Books and Semantris are both the first practical implementations of the Universal Sentence Encoder technique. This method was recently outlined in a paper published by a team of researchers at Google that includes world-renown author and futurist Ray Kurzweil.

Conceptually, the Universal Sentence Encoder technique focuses on representing language in a vector space using vectors that represents large text sections such as sentences or paragraphs vs. the traditional models that focused mostly on word vectors.

Source: towardsdatascience.com

Tags :
Share :
comments powered by Disqus

Related Posts

Artificial Neural Nets Grow Brainlike Navigation Cells

Artificial Neural Nets Grow Brainlike Navigation Cells

Having the sense to take a shortcut, the most direct route from point A to point B, doesn’t sound like a very impressive test of intelligence. Yet according to a new report appearing today in Nature, in which researchers describe the performance of their new navigational artificial intelligence, the system’s ability to explore complex simulated environments and find the shortest route to a goal put it in a class previously reserved for humans and other living things. The surprising key to the system’s performance was that while learning how to navigate, the neural net spontaneously developed the equivalent of “grid cells,” sets of brain cells that enable at least some mammals to track their location in space.

Read More
AI Can Generate ‘Doom’ Levels Now

AI Can Generate ‘Doom’ Levels Now

Researchers recently successfully trained neural networks to generate level maps for Doom that, they report in a paper published to the arXiv preprint server in April, “proved to be interesting” to play. The work was carried out by researchers from the Polytechnic University of Milan and used Generative Adversarial Networks, a recent innovation in the field of deep learning.

Read More
The fall of RNN / LSTM

The fall of RNN / LSTM

It is the year 2014 and LSTM and RNN make a great come-back from the dead. We all read Colah’s blog and Karpathy’s ode to RNN. But we were all young and unexperienced.

Read More