RT @risi1979: We're excited to announce the first Evolving Self-organisation workshop at @GeccoConf! Submission deadline: March 26, 2025…

RT @Clement_MF_: Can we study gene regulatory networks from a behavioralist perspective, as "agents" navigating in transcriptional space?…

RT @risi1979: Ever wish you could coordinate thousands of units in games such as StarCraft through natural language alone? 🗣️ We are exci…

@OxfordDPAG @ruimcosta @NatureComms That's the wrong Rui Costa @somnirons ...

RT @RichardMCNgo: Felix Hill was a kind and brilliant man (and my former teacher and colleague) who committed suicide after struggling with…

RT @douwekiela: I’m really sad that my dear friend @FelixHill84 is no longer with us. He had many friends and colleagues all over the world…

@tunguz Increasingly, people in academia don't know either...

RT @jeffrey_bowers: Multiple studies show that ANNs can learn languages relying only on general learning algorithms when trained on a diet…

@Plinz Damn that's 0 of 3 on takes in one post. Poverty of stimulus isn't disproven by point 1. AI can do 3 (to a high degree), but can't answer (yet) how language appeared in the first place. The mind that learns language could/can also invent it. Nobody ever questioned 2.

@m_tangemann @bethgelab Great thanks! Unfortunately had to submit a camera-ready version for a workshop already, but will make sure to use this citation instead of the arXiv one for the full article.

@bethgelab Is there a citation for this that is not the preprint?

@FelixHill84 I think one obvious reason why Transformers are better is that, while short dependencies are more common, long ones can be important. Those "edge" cases may be the difference between a good and a bad model. Transformers have a mechanism to overcome this, RNNs don't

RT @jeffrey_bowers: Reviewers & editors more likely to accept articles that use methods that produce (false) estimates of DNN-brain similar…

@mpshanahan No? Seems quite surprising to me. Was there no better contribution to physics than an application of ideas from physics to a different field?

@kenneth0stanley Data is inadequate. We don't just passively observe data, we interact with a world

RT @chrx_h: ... 2 days later, without much having happened, new type of organisms suddenly appear everywhere. Most of them are kind of asle…

@kenneth0stanley @fchollet @risi1979?

RT @fchollet: Easy -- being immersed in an ever-changing, adversarial ecosystem requires you to be able to adapt to novelty on the fly. Th…

RT @risi1979: And it’s a wrap! Thank you everybody for a great #ALIFE2024 conference.

RT @ALifeConf: How to make neural networks more flexible and adaptive by @JoachimWinther #ALife2024