Smerity

13 ms·

61.
by Smerity 9y ago
After reading about ZeroVM I was incredibly excited - but then saw the last Git commit was in January 2015 =[ Is it considered "stable" or is it simply abandoned? Are there any successors? Near zero overhead VM acting on data stor
62.
by Smerity 9y ago
Capsule networks[1] have literally not been used on anything larger than what is considered a toy dataset for computer vision yet. Whilst there is potential there I'd argue it's too early to ascribe extreme levels of excitement to
63.
by Smerity 9y ago
As another Australian, our housing prices is broken for multiple reasons but foreign investment is a _major_ contributor. Credit Suisse research indicates that "foreigners are acquiring 25 per cent of newly completed supply in NSW and
64.
by Smerity 9y ago
Amazon offer an official AMI which comes preloaded with various deep learning frameworks: MXNet, TensorFlow, CNTK, Caffe/2, Theano, Torch and Keras. For the P3 (Volta V100) instances you'll want to ensure you use an AMI preloaded
65.
by Smerity 9y ago
Honestly, I didn't spend enough time delving in to the Python overhead, especially in terms of the framework. Most of it would be an issue of my own causing however rather than the framework's. The original code I wrote was never
66.
by Smerity 9y ago
The P3 instances are the first widely and easily accessible machines that use the NVIDIA Tesla V100 GPUs. These GPUs are straight up scary in terms of firepower. To give an understanding of the speed-up compared to the P2 instances for a re
67.
by Smerity 9y ago
Whilst it might be a general concept, the most recent time I heard your paraphrased notes on warm introductions was Marc Andreessen at Startup School. Both the video and the transcript (with the warm introduction question first) are availab
68.
by Smerity 9y ago
Facebook are getting straight up annoying when it comes to their ads. The worst for me is their in feed video player. When an ad appears on a video, you can't like, comment, or share the video[1]. You're going to break my flow of
69.
by Smerity 9y ago
In practice, images are not particularly large and a batch of them would easily fit on a single GPU. What's more common is either (a) performing the forward and backward passes on 4 GPUs where each GPU has its own batch, then collectin
70.
by Smerity 9y ago
Hey Aidan, As noted, it's not a criticism of a reading list tailored for you, I simply feel it's a really confusing reading list given the likely audience that will arrive there from Hacker News and the title. I also don't th
71.
by Smerity 9y ago
With all due respect, this is a quite random reading list, and appears more a result of the 'if "deep learning" in post.title: post.upvote()' trend on Hacker News ... I will pick on two under the "Classics" sec
72.
by Smerity 9y ago
I presume you're already familiar with computing the numerical and analytical Jacobian[1][2] and just wishing for a better way? :) They're memory intensive as all hell and pretty finicky but at least it's something. I'll
73.
by Smerity 9y ago
As an extreme example it actually brought me a bit of hope. No source code was released but the "peer review" via arXiv, Twitter, and other various channels ended up bringing the story to a close. I'd like to imagine it'
74.
by Smerity 9y ago
+1. The "well-tuned proposed model and not-very-well-tuned baseline" is something I feel nearly every researcher is guilty of, including myself :) It's especially pronounced however when people compare to a baseline from pape
75.
by Smerity 9y ago
Agreed. Admittedly the paper that you're speaking about was on the extreme level of "bugged". The results were beyond stellar, attracting a great deal of interest, and researchers who are usually very reserved quickly pointed
76.
by Smerity 9y ago
To extend on what was written in the article, reproducibility is difficult in science generally but can be insanely difficult for machine learning. The key insight I've found over the years in the field, especially when applied to deep
77.
by Smerity 9y ago
This is all doable today. As an example, check out some bird photos[1] from the Visual Genome[2] project that are similar to your examples. I selected the photos and hosted them on Imgur in hopes we don't kill Visual Genome with traffi
78.
by Smerity 9y ago
As an Australian now living in San Francisco paying $55 per month for 100/100 (WebPass, which would be the same price but 500/500 or 1000/1000 if I moved to a newer building), I have the exact same feeling. When I journey bac
79.
by Smerity 10y ago
The consensus in the machine learning / deep learning research communities is that this is a colourful "nothing" visualization. Little to no information is conveyed, it doesn't represent the architectures in any sane way
80.
Uber’s new head of its AI labs has stepped down from his role(recode.net)
5 points by Smerity 10y ago|0 comments
81.
Stop saying DeepCoder steals code from StackOverflow(smerity.com)
2 points by Smerity 10y ago|0 comments
82.
by Smerity 10y ago
Disclosure: I'm one of the co-authors of the QRNN paper (James Bradbury, Stephen Merity, Caiming Xiong, Richard Socher) produced by Salesforce Research. There are many interesting advances that Deep Voice paper and implementation make
83.
by Smerity 10y ago
To some degree that's what happens. Internet Archive and Common Crawl use WARC files and deleting an individual entry or set of entries from them can be markedly difficult. It's easiest to mark it with some manner of tombstone for
84.
by Smerity 10y ago
Strongly agreed - Google's Project Zero helped them immensely and without them it would have continued to grow worse. The length of time this went for was already disastrous. From the Cloudflare blog post, "The earliest date memor
85.
by Smerity 10y ago
From the parent thread: The caches other than Google were quick to clear and we've not been able to find active data on them any longer. ... I agree it's troubling that Google is taking so long. That's really the c
86.
by Smerity 10y ago
There's a share button on each of the "tweets" at least on my screen. The direct link to the Sierpinski triangle inspired animation you're noting is below: https://www.dwitter.net/d/676
87.
by Smerity 10y ago
To note, they were also forced to delete the collected data, though the insights they've already extracted / profits from data they've already sold may offset both the point and the $2.2 million fine.
88.
by Smerity 10y ago
Extreme paper tldr - Humans usually construct neural network components and the graph of how they fit together by hand. This work sets up a "controller" neural network that constructs two core components in many neural networks, a
89.
by Smerity 10y ago
I am not highly clued in to XLA as it's new, quite experimental, and most honestly I've just not looked at it in detail. Given XLA provides compilation, JIT or ahead of time, it doesn't really (yet) factor in to the dynamic g
90.
by Smerity 10y ago
Torch was never written as a static graph computation framework. Torch was/is more a tensor manipulation library where you are executing the individual operations step by step and the graph can be tracked and constructed incrementally

More ›