President & Co-Founder @OpenAI
Joined July 2010
- Tweets 4,815
- Following 65
- Followers 498,999
- Likes 1,386
Here's the OpenAI Five technology applied to a physical robot: blog.openai.com/learning-dex…!
1
7
Great work! Congrats to the DeepMind StarCraft 2 team.
1/3 Really proud to share #AlphaStar with the world - a milestone for game-playing AI, and the first system to beat a top pro at #StarCraft2. Massive thanks to @Blizzard_Ent, @Liquid_MaNa & @LiquidTLO for being part of this great journey!
Blog + Replays: deepmind.com/blog/alphastar-…
3
17
1
184
Looking for professional Dota teams to playtest the latest version of OpenAI Five. Significant progress over the past few months. Email brooke@openai.com if interested.
36
111
12
786
A step towards showing that deep learning doesn’t need to be considered a mysterious art, but can be systematically studied and understood:
We can now measure a priori how well a deep learning model will parallelize during training. Super pumped that this work uses Dota as a research platform:
3
7
37
We can now measure a priori how well a deep learning model will parallelize during training. Super pumped that this work uses Dota as a research platform:
Announcing the first finding from our "Science of AI" effort: a simple statistical metric that lets us predict the parallelizability of neural network training on a wide range of tasks: blog.openai.com/science-of-a…
14
2
106
Today is Learning Day on our robotics team:
Thursday is a learning day at robotics@openai - reading papers, reimplementing toy examples, getting to know new engineering tools. We keep motivating each other by posting on slack what we have learned. Moreover, I wear special pants for every Learning Day.
2
5
69
Thanks for having me!
TWiST Live! @OpenAI Co-founder @gdb on rapid acceleration toward artificial general intelligence: deep learning, death of Moore's Law, risks of malice, competition & unintended consequences in the race to build AGI benefiting humanity-@jason @wilsonsonsini itunes.apple.com/us/podcast/…
2
8
35
These days The Matrix really is everywhere.
When designing a deep learning model, whenever a decision must be made, the answer is simple: multiply your data by a weight matrix. For example, in an LSTM, need to forget some data? Just multiply by a weight matrix to get the forget gate. The system will learn what you meant.
2
13
1
54
Our Fellows program turns people with software experience into ML practitioners, helping them make real contributions within only a small number of months. The latest example:
Excited to share the work I've done during my @OpenAI Fellowship! Using a new procedurally generated environment called CoinRun, we measure how well trained agents can generalize to new environments: blog.openai.com/quantifying-…
2
6
58
This is our third major attempt in the past two years (Universe, Retro Contest) to develop a platform for RL generalization. Each time, we’ve made the task easier — but more focused on the core generalization challenge. Already seeing promising results on CoinRun.
3
7
42
Today’s RL algorithms are great at exploiting a particular environment, terrible at using that knowledge in new situations. Here’s a new environment which is already helping us understand why, and which may help develop RL algorithms that generalize:
We’re releasing CoinRun, an environment generator that provides a metric for an agent’s ability to generalize across new environments - blog.openai.com/quantifying-…
8
64
1
196
Want to work on robotics? We're hiring. Looking for a software engineer with robotics experience who is familiar with writing code for the whole hardware stack.
The robotics team @OpenAI aims to build a general purpose robot as means toward AGI. Contact me if you'd like to help! We're especially looking for people who want to write software for hardware on all levels of abstraction - firmware, driver, grpc, python wrappers.
16
67
When designing a deep learning model, whenever a decision must be made, the answer is simple: multiply your data by a weight matrix. For example, in an LSTM, need to forget some data? Just multiply by a weight matrix to get the forget gate. The system will learn what you meant.
5
31
3
265
Applications for the second cohort of OpenAI Scholars are due by end of day tomorrow. We have an awesome group of mentors lined up; if you are from an underrepresented group in machine learning and want to get into ML, please apply! Details: blog.openai.com/openai-schol…
18
1
74
In 2018, OpenAI grew from about 40 full-time employees to over 80. Looking for recruiters (no recruiting experience needed) to help us continue to find new team members that make OpenAI an even better place for our existing people to work: jobs.lever.co/openai/5512201…
3
15
108