PaddlePaddle

@tachyeonz : PaddlePaddle (PArallel Distributed Deep LEarning) is an easy-to-use, efficient, flexible and scalable deep learning platform, which is originally developed by Baidu scientists and engineers for the purpose of applying deep learning to many products at Baidu.

Read More

Connect On:
Twitter :@tachyeonz

Human motion patterns learnt with conceptor-controlled neural network

@tachyeonz : A recurrent neural network learnt to re-generate a variety of motion patterns. The patterns were each defined by joint angle trajectories with 61 degrees of freedom in total.

Read More

Connect On:
Twitter :@tachyeonz

How to build and run your first deep learning network

@tachyeonz : Experiment with deep learning neural networks with Getting Started with Deep Learning using Keras and Python, an Oriole Online Tutorial by Mike Williams. When I first became interested in using deep learning for computer vision I found it hard to get started.

Read More

Connect On:
Twitter :@tachyeonz

Understanding LSTM and its diagrams

@tachyeonz : I’m not better at explaining LSTM, I want to write this down as a way to remember it myself. I think the above blog post written by Christopher Olah is the best LSTM material you would find. Please visit the original link if you want to learn LSTM. (But I did create some nice diagrams.)

Read More

Connect On:
Twitter :@tachyeonz

AI challenge in 78 lines (top 5%)

@tachyeonz : Tron is a multiplayer game in which 2-4 players play a snake-like game. Every step one makes, the board gets filled by one block. Whenever a player has nowhere to go, they lose. Last man standing. See a video below of 4 bots in action on Codingame:

Read More

Connect On:
Twitter :@tachyeonz

PathNet: A Modular Deep Learning Architecture for AGI

@tachyeonz : PathNet is a new Modular Deep Learning (DL) architectures that highlights the latest trend in DL research to meld Modular Deep Learning, Meta-Learning and Reinforcement Learning into a solution that leads to more capable DL systems.

Read More

Connect On:
Twitter :@tachyeonz

5 Free Courses for Getting Started in Artificial Intelligence

@tachyeonz : Interested in artificial intelligence? Don’t know where or how to start learning? Nothing substitutes rigorous formal education, but that’s not an option for everyone for a whole host of reasons.

Read More

Connect On:
Twitter :@tachyeonz

Astronomers explore uses for AI-generated images

@tachyeonz : AI-generated images of galaxies (left, lower of each pair) and volcanoes. Volcanoes, monasteries, birds, thistles: the varied images in Jeff Clune’s research paper could be his holiday snaps. In fact, the pictures are synthetic.

Read More

Connect On:
Twitter :@tachyeonz

The United States Of Artificial Intelligence Startups

@tachyeonz : Startups in 33 US states are applying machine learning algorithms across industries including marketing, healthcare, and retail. Deals to AI startups reached a record high last year, from 160 deals in 2012 to 658 in 2016.

Read More

Connect On:
Twitter :@tachyeonz

Listening to bots

@tachyeonz : After spending much of 2016 knee deep in the world of bots and conversational interfaces I thought it would be interesting to post what I have learnt and what I’m thinking about in 2017. We backed or built thirteen companies at betaworks in the bot, or conversational tools space last year.

Read More

Connect On:
Twitter :@tachyeonz

The Periodic Table of AI

@tachyeonz : This is an invitation to collaborate.  In particular, it is an invitation to collaborate in framing how we look at and develop machine intelligence. Even more specifically, it is an invitation to collaborate in the construction of a Periodic Table of AI. Let’s be honest.

Read More

Connect On:
Twitter :@tachyeonz

MIT’s New AI System: How It Learns By Surfing The Internet

@tachyeonz : Researchers from the Massachusetts Institute of Technology recently presented their new artificially intelligent system that can fill the information gap itself by surfing the Internet.

Read More

Connect On:
Twitter :@tachyeonz