Kyle (@darsnack) header
Kyle (@darsnack) profile picture

NeuroAI Scholar @CSHL | EE PhD @UWMadison | Comp. Eng. / math @RoseHulman | CS + neuro + AI

Joined September 2009
Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)Photo by Kyle (@darsnack)
We present Cheese3D, a system to sensitively track 3D movement of the entire face in lab mice. First #preprint from the lab! biorxiv.org/content/10.1101/… 1/n
24
86
12
348
59,013
The capabilities of LLMs are now causing a resurgence in folks saying that humans are "statistical parrots" mindlessly predicting the next word based on what they've heard before. 5 minutes of observing a young child still learning to speak should cure you of this notion.
Replying to @fchollet
In general I've been sensing a new current deep learning maximalists recently, going from "our models can definitely reason" to "well our models can't reason, but neither can humans!"
19
68
4
474
A dramatic example of this is how we engineer randomness into computers. Our amazing pseudo random number generators are reliable, predictable and rely on deterministism all the way down to transistors whereas biology is random at all levels and becomes deterministic when needed
Humans as a collective can develop some incredibly sophisticated systems (your computer and the software running on it is proof of that), but those still remain orders of magnitude short of the superhuman sophistication of biological systems. For now, at least.
1
1
#NeuroAI: Could principles of embodied sensorimotor neuroscience unify and improve the various Self-Supervised Learning (SSL) methods? How could the brain self-supervise itself? We are happy to share our #NeurIPS2022 paper arxiv.org/abs/2210.09224 with @franz_scherr and Q. Guo🧵:
6
27
4
134
This is a joke, but also is a neat idea. On a per lab basis, if you can only publish one paper per year (unlimited pre-prints), what would you work on? This constraint kinda forces within a lab all the good behaviors that we aspire towards.
This Post is from an account that no longer exists. Learn more
2
White paper —Rallying cry for NeuroAI to work toward Embodied Turing Test ! Let’s overcome Moravec’s paradox: Tasks “uniquely” human like chess and even language/reasoning are much easier for machines than “easy” interaction with the world which all animals all perform.
Massive whitepaper just dropped on why neuroscience progress should continue to drive AI progress: arxiv.org/pdf/2210.08340.pdf Argues for an embodied turing test. Needed: real interdisciplinary people, shared platform(s), fundamental research
6
9
1
43
Sometimes you come across numbers that are so mind boggling that you have to share them. Some people in America pay $100 / Mbps that others pay $0.25 / Mbps (in the same city)! We desperately need a scientific corps in Congress. themarkup.org/still-loading/…
1
Scholars in all disciplines & countries, we invite you to sign this open letter to @WHOSTP calling for equitable access & participation in research for all Why we think you should sign ostp-letter.github.io/why.ht… Or go straight to the letter ostp-letter.github.io #OSTPLetter
5
198
78
200
It’s not evident that making everything the same or even making those repeated units low power is leads to lower energy overall! How the stuff happening at higher abstractions exploits the substrate is more important for efficiency. This is where heterogeneity is a clear win.
Hey AI and neuromorphic followers - The brain is not homogeneous. The brain's heterogeneity is far more important than the efficiency gains you get by making everything the same.
Kyle retweeted
IMPACT SUCCESS! Watch from #DARTMIssion’s DRACO Camera, as the vending machine-sized spacecraft successfully collides with asteroid Dimorphos, which is the size of a football stadium and poses no threat to Earth.
7,424
47,792
12,456
228,476
It's #FreeMapMonday! Re-tweet and follow us for a chance to win the Satellite map of Hailey, ID! #USGSStore (U.S. residents only)
15
929
11
403
.@sarahookr coined the term "hardware lottery" to describe what happens when a particular algorithm takes off because it runs well on some new kind of hardware. What about when the hardware is designed to further cement the dominance of that algorithm? A hardware moat?
Ever since the V100, Nvidia has been cramming more and more "tensor cores" into each GPU generation. But what *are* tensor cores? How can you use them to accelerate deep learning models by >10x? And ... why does their existence make me somewhat sad :( (1/9)
12
12
95
I made this scrollytelling simulation of balanced excitation and inhibition in biological neural networks last year. JS+observable+React+Idyll. Everything runs live in the browser. It's the most in the weeds I've ever gone into JS balanced-ei.netlify.app
5
27
201
So it turns out bird's magnetic compass might be.. in their eyes? It boggles the mind what evolution can cook up scientificamerican.com/artic…
5
103
14
580
Can we please stop with the reimbursement system for grad students? I’m owed $7537.07 in reimbursements. This is more than 20% of my yearly income pre-tax. Some of these expenses occurred MONTHS ago. Expecting people to have this kind of money on hand is absurd
38
311
63
3,777
A cordial first interaction with Meta's GPT-3 clone.
11
37
20
379