PhD student at @GatsbyUCL and @ELSCbrain. Studying the mechanisms behind of intelligence 🧠🤖
Joined April 2016
- Tweets 10
- Following 164
- Followers 115
- Likes 5
Photos and videos
Jan Bauer retweeted
17 Sep 2023
Couldn't agree more: we still don't have good solutions for scalable abstraction/concept learning from "raw" data (be it language, vision, or other sensory data). Solving this would likely unlock important new capabilities.
17 Sep 2023
For anyone interested in future LLM development
One of the bigger unsolved deep learning problems: learning of hierarchical structure
Example: we still use tokenizers to train SOTA LLMs. We should be able to feed in bits/chars/bytes and get SOTA
Related: larger context window
4
15
117
29,950
Jan Bauer retweeted
The longer people are in academia, the more they realize that when reading papers it's best to ignore Intro, Discussion etc. and just look at Methods and Results
journals.plos.org/plosone/ar…
195
1,115
5,078
961,731
Jan Bauer retweeted
25 Jul 2023
We often discuss if the brain does something like gradient descent. Here is me discussing the issue for @OpenNeuroMorph. I focus on exposing the weaknesses of the hypothesis as well. Should be particularly useful for newly interested people. youtube.com/watch?v=E5hATeCZ…
3
25
83
8,805
Jan Bauer retweeted
21 Apr 2023
Ich bin @rezomusik Fanbase. Heute gemeinsames Video gedreht. Die Antwort auf Philip A… Kommt bald.
649
217
9,025
1,110,518
Jan Bauer retweeted
28 Jan 2023
never seen a graphic so wrong, but so useful
23
106
1,057
123,612
Jan Bauer retweeted
3 Jan 2023
💯 reminds me of MAML meta-learning (arxiv.org/abs/1703.03400) where the objective is to find weights of a network such that any new task finetunes fast. In Software 1.0 land, equivalent is writing code such that any new desired functionality is simple and doesn't need a refactor.
7
10
141
33,874
