brzozowski

8 ms·

31.
by brzozowski 6y ago
Interesting, thanks for the pointer Adam!
32.
Datascript: An immutable in-memory database and Datalog query engine in Clojure(github.com)
2 points by brzozowski 6y ago|0 comments
33.
Foundations of Computational Linguistics in Clojure(foundations-computational-linguistics.github.io)
2 points by brzozowski 6y ago|0 comments
34.
by brzozowski 6y ago
I have updated the WL(1) implementation to iterate until fixpoint termination, and mentioned the failure case. Thank you for reading carefully!
35.
by brzozowski 6y ago
> Have you thought of applying this to web-stacks? Yes, it seems like a good place to start. I was thinking of writing a PoC interpreter for a subset of the JS syntax. Thanks for the suggestion!
36.
by brzozowski 6y ago
Definitely the product of an over-excited grad student who’s fallen in love with these ideas for the first time. Thanks for reading, I hope you were able to get something from it (if not what to avoid)!
37.
by brzozowski 6y ago
Yes, I was originally inspired by GraalVM’s work! They were one of the first to realize a practical implementation of Futamura’s ideas. I briefly discuss it towards the end: http://breandan.net/2020/06/30/grap
38.
by brzozowski 6y ago
Sure, I made plenty of bad predictions too! I have written about some of the harmful effects of trust and bias in automation, feel free to check it out: https://breandan.net/2017/02/02/trust-in-automation&#x2F
39.
by brzozowski 6y ago
Very cool, thank you! Link was broken, here is the full reference: https://www.cs.utexas.edu/~pingali/CS395T/2009fa/papers/ferr...
40.
by brzozowski 6y ago
It has been rebranded a number of times, from automatic differentiation to algorithmic differentiation to differentiable programming. Takeo Imai wrote a blog post about the origins and history of the term and its modern usage: https:/
41.
by brzozowski 6y ago
> It seems like the overall goal is to provide some intuition for how automatic differentiation & program synthesis might be implemented efficiently using matrices. Right! I wish I had emphasized that theme a little more. This is bas
42.
by brzozowski 6y ago
> I hope you mean by parallel processing, on a single machine. Not necessarily! Parallelism in the more general sense. We can parallelize matrix multiplication and there is good research on distributing graph algorithms across multiple m
43.
by brzozowski 6y ago
Thanks for your comments! I enjoyed reading about STKs, which I had not heard about before. I know lots of people are interested in applying kernel methods to graphs, and it seems like a important area of research. As your post discusses, o
44.
by brzozowski 6y ago
Hi Dan, thanks for taking the time to read this and leave your feedback. I'll try to address your remarks one-by-one. > I found the frequent self-quotes odd and the self-congratulatory introduction more odd. Thank you for the candid
45.
by brzozowski 6y ago
Thanks, this is helpful feedback. I was thinking about this, and where it comes from in my writing. I’m not going to erase these details, because it documents who I am, and my frame of mind. If you leave with a poor opinion of the writer be
46.
by brzozowski 6y ago
Thanks for your feedback! I like the WL algorithm for its simplicity, but agree there are specific cases it does not handle well. It is meant to illustrate a simple message passing algorithm, and is not a particularly efficient implementati
47.
by brzozowski 6y ago
I would say that’s becoming increasingly plausible, but it’s not exactly what the authors show in this paper. In order to translate from Datalog queries, you would need to show how to encode arbitrary propositional formulae as a graph reach
48.
by brzozowski 6y ago
Yes, “program synthesis” or “program induction” are the broader topics, which span automata theory, type theory and (more recently) representation learning. Prior work has relied on classical algorithms like proof search, SMT solvers and le
49.
Learning Graph Structure with a Finite-State Automaton Layer(arxiv.org)
92 points by brzozowski 6y ago|5 comments
50.
Learning Differential Equations That Are Easy to Solve(arxiv.org)
4 points by brzozowski 6y ago|1 comments
51.
Intuitive physics, planning, and problem-solving in brains, minds and machines(youtu.be)
1 points by brzozowski 6y ago|0 comments
52.
by brzozowski 6y ago
Related article by Russell O’Connor about copyright assignment I recently stumbled onto: http://r6.ca/blog/20110930T012533Z.html
53.
Structural Language Models of Code(arxiv.org)
2 points by brzozowski 6y ago|0 comments
54.
Introduction to Program Synthesis(people.csail.mit.edu)
3 points by brzozowski 6y ago|0 comments
55.
by brzozowski 6y ago
Have you tried the TeXiFy IDEA plugin? It's well-maintained and has saved me a bunch of time. https://github.com/Hannah-Sten/TeXiFy-IDEA https://plugins.jetbrains.com/plugin/9473-texify-idea