Profile avatar
neuralreckoning.bsky.social
Computational neuroscientist at Imperial College. I like spikes and making science better (Neuromatch, Brian spiking neural network simulator, SNUFA annual workshop on spiking neurons). 🧪 https://neural-reckoning.org/ 📷 https://adobe.ly/3On5B29
1,475 posts 3,238 followers 331 following
Regular Contributor
Active Commenter
Posts 14 Comments 36

Have you tried building neuronal models with #DendroTweaks? It’s a Python toolbox we made to help you visualize and customize single-neuron models with active dendrites. And it's just got a major update! ✨ Here's what's new: 🧵

submitted 6 hours ago • 1 comment

🧠 How do transformers learn relational reasoning? We trained small transformers on transitive inference (if A>B and B>C, then A>C) and discovered striking differences between learning paradigms. Our latest work reveals when and why AI systems generalize beyond training data 🤖

submitted 11 days ago • 2 comments

Are there useful or interesting ways to use LLMs other than prompting them? I feel like compressing all the text in the world via a hierarchically structured statistical model is probably useful, but that we're using it in a way that is unlikely to do what we'd hope.

submitted 2 days ago • 1 comment

My 3 year old made me a delicious "magic cake" for father's day. Yummy.

submitted 2 days ago • 1 comment

I just started listening to the Night Science podcast and it's brilliant. All about the very different creative processes scientists use. This episode seems like a good place to jump in.

submitted 2 days ago • 1 comment

Well this was creepy.

submitted 3 days ago • 1 comment

I'm so confused. Today is actually Trump's birthday and he's having a parade, and at the same time it's the birthday parade (not actually his birthday) of the king of England, and there's a No Kings demonstration against Trump, but not Charles. Is that right? And these things are unrelated? 🤯

submitted 3 days ago • 2 comments

Haven't read the paper yet, but I'm happy to see this. It's essentially the reason why I've been arguing for post publication peer review. The people best placed to review well are the people actually using the results.

submitted 3 days ago • 0 comments

Ok, development and ethics peeps, I have a testing dilemma for you: A kid comes to the lab to take part in a neuroimaging study (EEG, MRI, cognition) with their mom. Consent is signed (only 1 legal guardian signature is required by ethics). The parents are divorced, the dad freaks out and (1)

submitted 3 days ago • 1 comment

Just saw an advert for "quantum energy coaching" and wondering if a quantum of energy is more or less than a homeopathic dilution?

submitted 4 days ago • 0 comments

Is it true you use only 10 bits / s of your brain? @andpru.bsky.social and I provide a motor systems perspective on a proposed speed limit for human information processing. Out today in @natneuro.nature.com: www.nature.com/articles/s41...

submitted 4 days ago • 8 comments

If you're sorting your Pocket saves, this Python code might be useful. Export your Pocket data to csv file from Pocket website, then run this code and it will generate a single HTML page with all your unread links sorted by date and grouped by tag. gist.github.com/thesamovar/8... #pocket #firefox

submitted 5 days ago • 0 comments

The Cognitive Computational Neuroscience conference is in Amsterdam this year, but you don't have to go that far to attend. Local meetups/watch parties are being organized all over, including the one I'm hosting in New York! Check them out and register (free!) here 👇 #neuroskyence #compneurosky

submitted 6 days ago • 3 comments

We are organizing a workshop at CNS*2025 in Florence, Italy. It will be about structure, dynamics and function in neural networks. Please see the homepage for more details. sites.google.com/view/cns2025...

submitted 21 days ago • 3 comments