Dataflow matrix machines (by Anhinga anhinga) (
dmm
) wrote
2020
-
06
-
29
12:46 pm
Applied Category Theory Conference: July 5-10
act2020.mit.edu/
Online, free. July 5 - tutorial day:
act2020.mit.edu/#tutorialday
Flat
|
Top-Level Comments Only
no subject
dmm
2020-07-05 04:42 am (UTC)
(
link
)
https://en.wikipedia.org/wiki/String_diagram
no subject
dmm
2020-07-05 06:38 pm (UTC)
(
link
)
The slides for Tutorial #3, "The Yoneda lemma in the category of Matrices":
http://www.math.jhu.edu/~eriehl/matrices.pdf
Now all my objections to the use of categories in programming come back to me.
Categories are difficult. At least, this tutorial, aiming to simplify difficult things, is still too difficult.
A typical programming activity should not be difficult, a typical programming activity should be easy.
no subject
dmm
2020-07-06 11:37 am (UTC)
(
link
)
First keynote:
Main reference: Elements of Petri nets and processes
http://arxiv.org/abs/2005.05108
Graph paper (Joachim Kock): Graphs, hypergraphs, and properads
http://arxiv.org/abs/1407.3744
no subject
dmm
2020-07-06 12:50 pm (UTC)
(
link
)
Further remarks at the first session:
"Competently playing with categorical patterns seems to be a very reasonable goal for an AI system; I hope a good research group would work on this at some point." (
https://twitter.com/ComputingByArts/status/1280117144508866560
)
Last talk of the first session:
https://arxiv.org/abs/2005.05293
no subject
dmm
2020-07-06 04:36 pm (UTC)
(
link
)
The whole "categorical probability" field is very interesting and active, with remarkable new developments.
The keynote of the session dedicated to it was very cool (second session of the first day).
Also, with probability, moving to categories actually might be making things simpler ;-)
no subject
dmm
2020-07-08 04:11 pm (UTC)
(
link
)
I am behind on my comments; hopefully I'll rectify that. There were some very interesting moments.
Meanwhile, it seems that a lot of it will remain on YouTube, so it would be possible to revisit the most interesting talks...
Right now, the second session of Wed keynote on
https://github.com/AlgebraicJulia/Catlab.jl
no subject
dmm
2020-07-08 04:45 pm (UTC)
(
link
)
Wed industry session:
From David Spivak to Everyone: 12:41 PM
• Alan Ransil (Protocol Labs)
• Ryan Wisnesky (Conexus)
• Jelle Herold (Statebox)
• Steve Huntsman (BAE)
• Arquimedes Canedo (Siemens Corporate Technology)
• Ilyas Khan (Cambridge Quantum Computing)
• Brendan Fong (Topos Institute)
• Alberto Speranzon (Honeywell)
A super-interesting session so far (at half-point).
Edited
2020-07-08 17:20 (UTC)
7 comments
Post a new comment
Flat
|
Top-Level Comments Only
[
Home
|
Post Entry
|
Log in
|
Search
|
Browse Options
|
Site Map
]
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
no subject
no subject
Now all my objections to the use of categories in programming come back to me.
Categories are difficult. At least, this tutorial, aiming to simplify difficult things, is still too difficult.
A typical programming activity should not be difficult, a typical programming activity should be easy.
no subject
Main reference: Elements of Petri nets and processes http://arxiv.org/abs/2005.05108
Graph paper (Joachim Kock): Graphs, hypergraphs, and properads http://arxiv.org/abs/1407.3744
no subject
"Competently playing with categorical patterns seems to be a very reasonable goal for an AI system; I hope a good research group would work on this at some point." ( https://twitter.com/ComputingByArts/status/1280117144508866560 )
Last talk of the first session: https://arxiv.org/abs/2005.05293
no subject
The keynote of the session dedicated to it was very cool (second session of the first day).
Also, with probability, moving to categories actually might be making things simpler ;-)
no subject
Meanwhile, it seems that a lot of it will remain on YouTube, so it would be possible to revisit the most interesting talks...
Right now, the second session of Wed keynote on https://github.com/AlgebraicJulia/Catlab.jl
no subject
From David Spivak to Everyone: 12:41 PM
• Alan Ransil (Protocol Labs)
• Ryan Wisnesky (Conexus)
• Jelle Herold (Statebox)
• Steve Huntsman (BAE)
• Arquimedes Canedo (Siemens Corporate Technology)
• Ilyas Khan (Cambridge Quantum Computing)
• Brendan Fong (Topos Institute)
• Alberto Speranzon (Honeywell)
A super-interesting session so far (at half-point).