This is a teriffic article and I can't wait for the next few in the series (I found it via Data Science Weekly but I've now subscribed). It's a retrospective from the first day of the week-long Cosyne 2026 conference and breaks down into three main parts: a short discussion of open databases, introduction to a 4 hour (!) session on ways to compare neural signals, and a long discussion of the keynote from Chris Olah, co-founder of Anthropic. There are many gems in here, some things that reinforced my previously held views, and others than challenged them. One thing that matters to me: "we find that networks tend toward distributed representations and mechanisms, which make understanding both artificial and biological networks a pain, equally... , the most natural computational unit of the neural network – the neuron itself – turns out not to be a natural unit for human understanding. This is because many neurons are polysemantic: they respond to mixtures of seemingly unrelated inputs."
Today: Total: [Direct link] [Share]
