Latent

PhD student ML, Inverse problems.

[Latent @RepresenterTh]

@SakanaAILabs Does each block have its own optimizer ?

[Latent @RepresenterTh]

@fchollet This is such nonsense because it assumes completeness of maths and posteriority of understanding. Godel an esteemed logician and mathematician would not like such a take.

[Latent @RepresenterTh]

@astro_jaz First 3 seasons

[Latent @RepresenterTh]

@ProfTomYeh and prelu generalize the activation by making \alpha learnable and scalable.

[Latent @RepresenterTh]

@fchollet Nonsense

[Latent @RepresenterTh]

@antigravity The rate limits are ridiculous and if you want people to use Gemini update the model. 3.1 is just bad.

[Latent @RepresenterTh]

@francoisfleuret They'll have a comeback if nested learning makes it.

[Latent @RepresenterTh]

@jeankaddour Stronger penality for the model in blue ( larger weight decay ) or the model in black is just way bigger explaining the rapid descent after warmups.

[Latent @RepresenterTh]

@johannesack Replace the KL with maximum mean discrepancy (gretton) using an rbf kernel.

[Latent @RepresenterTh]

@fchollet Being a decimal and a limit of a series are not mutually exclusive.

[Latent @RepresenterTh]

@suchnerve Equality no, convergence yes. I. Maths two quantity cannot be equal if they don't share the same properties. 0.999 is not an integer.

[Latent @RepresenterTh]

@ai_for_success Anthropic and its models

[Latent @RepresenterTh]

@j_dekoninck AIME has to go

[Latent @RepresenterTh]

@emollick "Chinese open weight model remain 9 months behind" only of you trust useless and biased benchmarks funded by close source.

[Latent @RepresenterTh]

@jasondeanlee Nonsense

[Latent @RepresenterTh]

Intelligent systems are not rule followers (capacity). A system whose competence is entirely explained by reward maximization is not intelligent. An intelligent system does not need low level instructions to perform a task. An intelligent system cannot be goalless.

[Latent @RepresenterTh]

@WeLivetoServe @ylecun @demishassabis Wtf

[Latent @RepresenterTh]

@sytelus @demishassabis Not really.

[Latent @RepresenterTh]

@demishassabis There may be layers of this world and the universe that we do not know about because our intelligence is not general enough. Again this isn't worth debating because there's not a general consensus on these terms.

[Latent @RepresenterTh]

@demishassabis In any case human intelligence is the most general form of intelligence we know of. Yann may see general intelligence at a lower level. Something more fundamental that transcends domains. After all humans are highly optimized (by design) to operate in the physical world.