Progress towards (but not solved) - use of laws of physics as a prior - interal “world models” in RL - Nash equilib as learning (ie Tsetlin machine) - more general DNNs (ie Graph Networks) - better problem formulation (ie Neural ODE)
a year ago today I posted “deep learning: a critical appraisal”, asking whether DL might be approaching a wall, and be just one tool among many needed for AGI. hive mind: what progress has been made? what new limits have been identified? what other tools have been invented?
-
-
-
Best reference(s) for the first, please?
- 2 more replies
New conversation -
-
-
I liked this paper from NeurIPS: https://papers.nips.cc/paper/7948-end-to-end-differentiable-physics-for-learning-and-control … They put a deep learning wrapper around a differentiable physics engine and then can rapidly learn to fix the errors of the physics engine. They learn breakout in a few thousand steps.pic.twitter.com/KU3m3gmUHN
-
Authors: Belbute-peres, Filipe De A; Allen, Kelsey R; Smith, Kevin A; Tenenbaum,; Joshua B; Kolter, J Zico
End of conversation
New conversation -
-
-
In the shadow of the deep learning hype, Neuromorphic Computing is starting to get really interesting:https://twitter.com/samim/status/1077888732177543168 …
- 1 more reply
New conversation -
-
-
I think we discovered the limits of deep twitter debates :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Did you know about
#IEML, an artificial language with a *computable semantics* at the paradigmatic, syntagmatic and textual levels? It may be used as a knowledge representation framework for IA, among other possible uses. More: https://pierrelevyblog.com/my-research-in-a-nutshell/ … andhttps://pierrelevyblog.com/my-research-in-a-nutshell/the-basics-of-ieml/ …Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
My opinion (which everyone is entitled to) is the Semantic Web is true AI: it enables reasoning using logic. In doing so it exploits the tight connections between logic and language. How difficult do we have to make it?
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.