It misses some (see the overlay in the repo). And then it convolves instead of just histogramming it

GitHub - koerding/dopamineRaster: Digitization and modern replot of a classic dopamine raster figure Digitization and modern replot of a classic dopamine raster figure - koerding/dopamineRaster

github.com/koerding/dop...

Feel free to play with it x.com/kordinglab/s...

I use a kernel for smoothing…

Everyone knows this graph. I tried to clean it up/ make it modern. Attention some dataloss is unavoidable.

github.com/koerding/dop...

AI + Us featuring Konrad Kording

Come to a brewery to night, have a cold one, and talk ai with Bhuv Jain and yours truly. AI is awesome and it also sucks. Let’s talk about it. ai.upenn.edu/ai-us-featur...

We’ve got an exciting new thing to share! We have causal evidence (using TMR) that memory reactivation during sleep promotes abstract understanding of underlying structure, allowing transfer learning in a new domain with zero superficial feature overlap with the learned one.

The hardest part of science is posing the right question, not answering it.

But sometimes you can replace a technique for a question with another more rapidly scaling technique. Everything is now sequencing. In neuro an awful lot will be imaging.

I was just going to make sure everyone has seen this. Imaging technologies double in efficiency roughly every year. I foresee a future with a lot of imaging (somewhat sadly cause I am an electrode guy).

There would not be a molecular signature? Why not?

Stoked this is finally out! We ask: how can we simulate the brain from the bottom up? It's not sufficient to grab the connectome and wire it up in silico! We need 1) ultrastructure 2) (causal) calibration data 3) functional data. Then we can build a simulation compiler. 1/

Great to work w/ @kordinglab.bsky.social and team on this important roadmap for linking molecular/structural features to function. The future of precision treatments for brain disorders, neurotechnology advances, and an understanding of intelligence itself depend on science that bridges btw scales

Cause you made them!

@davidamarkowitz.bsky.social , Jordan Matelsky, Brett Mensh,
@patrickmineault.bsky.social , Andrew Payne, Joanne Peng, Xaq pitkow, Phillip Shiu, Gregor Shuhknecht, Sven Truckenbrodt, Josh Vogelstein, @eboyden3.bsky.social

Major collaboration with @antonarhipov.bsky.social @seanescola.bsky.social Davy Deng @galhaspel.bsky.social Michal Januszewski, Bobby Kasthuri, Nina Khera, Richie Kohman, @neurograce.bsky.social ,
Jeantine Lunshof, @adammarblestone.bsky.social,

In any case, our point is that compiling molecularly annotated ultrastructure into dynamics is timely. I should probably also mention the 3 page "There are many things that could go wrong" with this appendix.

Others believe having the right amount of micro data would enable better AI. Or simulations of whole nervous systems. While everyone has their own goals, the value of compilers would help everyone.

What would it enable? Entirely depends on whom you ask. Single cell modeling needs parameters, which we currently can't measure. Many physiology questions require micro-parameters. Many disease related questions require micro-data.

What does it require to produce such a compiler? A lot of paired data. e.g. careful synaptic physiology with subsequent imaging. And dendrites, axons, etc. Such work will be hard but majorly improve the value of incoming imaging data. For everyone in the field.

For now there is a wide gulf between imaging giving us structure and physiology giving us function. We argue that we need to build a compiler that translates the former into the latter.

The time is ripe for compilers because imaging is getting better and cheaper at an unbelievable speed - voxels per dollars is doubling every roughly 1.5 year, at rapidly improving quality. 3d-imaging becomes attractive for neuroscience.

Preprint out arguing that we should build the techology to translate (compile) molecularly annotated connectomes into dynamics. I think this is incredibly important. arxiv.org/abs/2603.25713

Til; When I was young all talks were full of deep thinking.

I don't think I would characterize the role of AI as such. We don't yet know, I think, what it will do to the distribution of money.

Sure. But how is that related to the article?

The key detail everyone’s getting wrong about AI and the economy Opinion: Konrad Körding and Ioana Marinescu from the University of Pennsylvania argue artificial intelligence will likely have a limited impact on jobs because of the realities of physical work

With @imarinescu.bsky.social I argue that the economy is bottlenecked by the physical, rendering anything resembling a singularity unlikely: www.transformernews.ai/p/the-key-de...

"We are watching two disciplines trade their worst habits. Neuroscience is mistaking benchmarked prediction for understanding, and machine learning is mistaking mechanistic language for mechanism. ..."

Read my critical take on this development ;)

Either way. Concluding about city design is pretty off the charts,