moshe

8 ms·

1.
A Discrete Model of the Euclidean Plane(moshelooks.github.io)
1 points by moshe 5y ago|0 comments
2.
by moshe 10y ago
I've never used PyTorch or Chainer, so maybe I'm wildly wrong, but here goes.. Dynamic batching requires you to predefine a set of "batch-wise" operations (node types for computation graphs). This works fine for eager ex
3.
by moshe 10y ago
Thanks, insightful questions. You're absolutely right about combinators and long-range dependencies. We have a special block type in the high-level API ( https://github.com/tensorflow/fold/blob/master&#x2F
4.
by moshe 10y ago
Please note that the GitHub page says "not an official Google product", rather than "project". An official Google product would be something like gmail.
5.
by moshe 10y ago
TensorFlow Fold provides a TensorFlow implementation of the dynamic batching algorithm (described in detail in our paper [1]). Dynamic batching is an execution strategy for computation graphs, you could also implement it in PyTorch or Chain
6.
TensorFlow Fold: Deep Learning with Dynamic Computation Graphs(research.googleblog.com)
195 points by moshe 10y ago|24 comments
7.
by moshe 18y ago
I have gotten plenty of great applicants - thanks guys! Please stop sending me your application info at this point...
8.
by moshe 18y ago
It is, actually.
9.
by moshe 18y ago
"Currently pursuing a BS, MS or PhD in computer science or a related technical field." is verbatim from the intern job listing ( http://www.google.com/support/jobs/bin/answer.py?answer=1262... ), and I suppose it is there because the target
10.
Common Lisp + Machine Learning Internship at Google (Mountain View, CA)
45 points by moshe 18y ago|16 comments
11.
by moshe 18y ago
> Intelligence is general and is capable of coping with > novel problems. AI therefore invovles building machines > which have that level of generality. > There is a bias when you think of "programs" to think of programs
12.
by moshe 18y ago
a) Not everyone in Google Research has a PhD, though of course there is a bit of a correlation (also, plenty of "research" takes place in general engineering) b) You do generally have to be very familiar with one of the Big 4 to get hired (
13.
by moshe 18y ago
Well, the difficulty with this is that program induction is in fact too general to be very interesting - to create a competition you would need to publish some test problems, or describe a special distribution of "natural programs" that y
14.
by moshe 18y ago
Well, sort of. Genetic programming with probabilistic modeling and sampling instead of crossover/mutation (i.e. it an estimation-of-distribution algorithm), where programs are represented in reduced (normal) form and which variables to use
15.
by moshe 18y ago
Thanks for the pointer. I was actually just looking at ACL the other day - I have a friend here at Google who is investigating integrating it with plop as a 20% project. I'm very excited that your work combines theorem-proving with learning
16.
by moshe 18y ago
Heh, thanks. moshe at google dot com was already taken, it was my second choice...
17.
by moshe 18y ago
None taken. Thanks!
18.
by moshe 18y ago
But it is! I really am lucky enough to get to do machine learning research at Google and code in Common Lisp! See http://research.google.com/ . Another way you can tell that this is an official Google project is the 'Google' label on the r
19.
by moshe 18y ago
Eventually, learn arbitrary programs from data, e.g. Input: (learn 'fib '(x) '(((1) 1) ((2) 1) ((3) 2) ((4) 3))) Output (defun fib (n) (if (< n 3) 1 (+ (fib (1- n)) (fib (- n 2))))) as well as stan
20.
by moshe 18y ago
Actually, I work full-time for Google Research (and have written all of the plop code, so far). Google happens to have sponsored a bunch of summer-of-code students through opencog.org, one of whom I supervised, but he coded in C++. I think