dennybritz

9 ms·

31.
by dennybritz 9y ago
Prediction Machines | DevOps/Backend/ML/Engineer | Tokyo, Singapore, Bangkok (REMOTE possible) | Competitive salary depending on location At Prediction Machines, we're applying Deep Learning and Reinforcement Learning te
32.
AI and Deep Learning in 2017 – A Year in Review(wildml.com)
6 points by dennybritz 9y ago|0 comments
33.
AI and Deep Learning in 2017 – A Year in Review(wildml.com)
3 points by dennybritz 9y ago|0 comments
34.
Mastering Chess and Shogi by Self-Play with General Reinforcement Learning(arxiv.org)
539 points by dennybritz 9y ago|270 comments
35.
77-year old Japanese creates paintings in Microsoft Excel(youtube.com)
8 points by dennybritz 9y ago|0 comments
36.
by dennybritz 9y ago
Prediction Machines | Backend/ML/DevOps Engineer | Tokyo, Singapore, Bangkok (REMOTE possible) | Very competitive salary depending on location At Prediction Machines, we're applying Deep Learning and Reinforcement Learning te
37.
Hacks for Training Deep RL Systems(github.com)
2 points by dennybritz 9y ago|0 comments
38.
MCMC Interactive Gallery(chi-feng.github.io)
2 points by dennybritz 9y ago|0 comments
39.
by dennybritz 9y ago
The other big "hardcoded" constraint is most likely the usage of the bot API. The API itself is complex with lots of functionality and I would assume that researchers were extremely smart about picking out just the right API calls
40.
by dennybritz 9y ago
I agree. The point I was trying to make is not whether it is impressive or not (it is impressive!), but that the general press hype about "AI breakthroughs that will soon kill us all and need regulation" is over the top and mislea
41.
by dennybritz 9y ago
Author here. I agree this is an assumption, but based on my experience it is very unlikely that this is trained on pixels. Training would've been orders of magnitude more expensive. If it really is trained on pixel input I would be s
42.
Hype or Not? Some Perspective on OpenAI’s DotA 2 Bot(wildml.com)
170 points by dennybritz 9y ago|125 comments
43.
Hype or Not? Some Perspective on OpenAI’s DotA 2 Bot(wildml.com)
2 points by dennybritz 9y ago|0 comments
44.
The Entrepreneur Life [video](youtube.com)
1 points by dennybritz 9y ago|0 comments
45.
Revisiting Unreasonable Effectiveness of Data in Deep Learning Era(arxiv.org)
3 points by dennybritz 9y ago|0 comments
46.
We haven't had a major new technology company in more than 10 years(vox.com)
34 points by dennybritz 9y ago|8 comments
47.
by dennybritz 9y ago
This has been said before by another commenter, but I'll say it again because I think it really is true: Most likely you will never regret taking the year off, but in a few years you probably WILL regret never taking that year off. &gt
48.
by dennybritz 10y ago
This is a good question and we should probably add this to the FAQ. tf.contrib.seq2seq is a low-level library that you can use to build seq2seq models; it is used internally by this project. The key difference here is that the google&#x2F
49.
by dennybritz 10y ago
Yes, these models can applied to a lot of non-NLP tasks. For example, I've seen seq2seq models applied to medical record prediction, program generation, etc. Noise removal seems like a good candidate.
50.
Show HN: A general-purpose encoder-decoder framework for Tensorflow(github.com)
54 points by dennybritz 10y ago|7 comments
51.
Engineering is the bottleneck in Deep Learning research(blog.dennybritz.com)
8 points by dennybritz 10y ago|0 comments
52.
by dennybritz 10y ago
A lot of recent research uses RL to "fine-tune" NLP models. A practical example would be Google's recently announced Machine Translation System ( https://arxiv.org/abs/1609.08144 ). It uses RL to directly
53.
by dennybritz 10y ago
Another reason may be that this is a pure "resource" post that doesn't make an argument or represents personal opinions that people could easily comment on. It's not a good basis to start a discussion, unlike many other
54.
Learning Reinforcement Learning, with Code, Exercises, and Solutions(wildml.com)
305 points by dennybritz 10y ago|21 comments
55.
RNNs in Tensorflow, a Practical Guide and Undocumented Features(wildml.com)
4 points by dennybritz 10y ago|0 comments
56.
Deep Learning for Bots Part Part 2 – Retrieval-Based Models in Tensorflow(wildml.com)
3 points by dennybritz 10y ago|0 comments
57.
by dennybritz 10y ago
These libraries are not mutually exclusive, they operate on different levels of abstraction. For example, you can use Keras with either Theano or Tensorflow. cuDNN is an interface to the hardware and is used internally by most libraries. Br
58.
Machines that dream: Understanding intelligence with Yoshua Bengio(oreilly.com)
3 points by dennybritz 10y ago|0 comments
59.
Stanford NLP Reading Group Schedule(nlp.stanford.edu)
2 points by dennybritz 10y ago|0 comments
60.
Deep Learning for Chatbots, Part 1 – Introduction(wildml.com)
9 points by dennybritz 11y ago|0 comments

More ›