I am a systems, and computational neuroscientist at MIT.
Primary research work is on the neurophysiology and computational modeling on visual attention, eye movements and executive functions, and the theory of neural networks.
Interests span math, science, history, politics, economics, and philosophy.
Our paper, "Stimulus representations in visual cortex shaped by spatial attention and microsaccades" was just published in the Proceedings of the National Academy of Sciences (PNAS).
https://www.pnas.org/doi/10.1073/pnas.2420704122
#Attention #VisualAttention #SpatialAttention #VisualCortex #Microsaccades #Saccades #EyeMovements #Representation #Dynamics #ActiveCognition #ActiveSensing #Decoding #V4cortex #ITcortex #Pulvinar #Neuroscience #PNAS
Thank you!
We are organizing and hosting the USA Memory Championship Organization's "Tournament of Memory Champions" at the Kresge Auditorium at MIT on November 10th(Sunday).
During the event, we will have short scientific talks about the benefits of mnemonic and cognitive training in health and in education.
The event is free and open to the public!
Please register here, and join us if you are in the Boston area:
https://www.eventbrite.com/e/the-mit-tournament-of-memory-champions-tickets-1042035515277
#HumanMemory #BiologicalMemory #Neuroscience #Biology #Mnemonics #CognitiveTraining #MIT #USAMC
Interested in the mysteries of human memory? Come check out the MIT Tournament of Memory Champions on November 10th at Kresge Auditorium. Organized by a few neuro-friends.
Wrote an essay on memory after a similar even in 2017. 👇
Sorry, have been off the social media bandwagon, just saw your message.
I don't fully understand what you mean by "simultaneous multiple meanings" here, but the different closed contour paths that one can take on the analytical image would give rise to different representations. Is that "meaning"(ful)? I don't know
something something... vector embedding, gazillion dimensions and neural manifolds mate! that is all there is.
use a few more loaded words, it obviously is on the path to sounding scientific.
don't forget, they have seismic sensing as well!
ugggh... that's what I meant. the bipole cells for grouping (von der Heydt style)!
@TonyVladusich
oooh... nice. Dipole cells!!
Thank you thank you. Will try to organize stuff from my end too!
Trying to straighten the curve and then hoping to recover/represent the curve!
Also, we have been dancing around Todd's thesis too much. Must sit and work through it!
Neurons being one difference. There are several other major differences, at the level of learning, network dynamics, functions, and behaviors too.
Ooh... great question! I don't remember the exact words I said, but maybe, something about the fact that:
1) I don't believe incorporating neurons or convolutions makes the network now tractable to solve/understand biological vision. I personally think using these networks as models of the world (brain) is a wild-goose chase.
2) Gradient descent/optimization approaches are quite removed from how humans and animals learn.
3) I spoke at length about gestalt psychology. Now it is vulgarized, by especially object recognition researchers (who can't seem to read anything outside of deep-nets) as "mid-level vision". We are nowhere near addressing those ideas with these models, and my suspicion is we won't get there by simply image computable deepnets.
So in summary: deep-nets are a remarkable engineering feat that work splendidly well in narrow domains, but are a red-herring for doing science.
I am increasingly realizing how what we (a few of us, say, vision psychophysics and neuroscience folks) find baffling is taken for granted by most people, i.e., the seeming perception of a high resolution and rich visual world.
Nowadays, when I meet most people, I do party tricks; like, "take a look at what's in front of you" then ask them a detail they would most certainly miss, or do the rule of thumb to show how limited (space variant) our vision is, not to mention how we automatically gain control across different lighting/color conditions and so on; to show how profoundly different our visual system is from existing computer vision systems from the very outset.
I was interviewed by The Economist's Babbage podcast on their series, "The science that built AI" last month. My hour long conversation was edited to about six minutes!
I am glad they edited/fit my conversation as taking the perspective that this big data, big compute driven deep-net approach is orthogonal to human/biological vision. And that, without incorporating biological principles (in this case, vision), autonomous visual navigation systems (i.e., self-driving cars) are unlikely and/or limited.
Unfortunately, the podcast requires a subscription to The Economist (I too had to access it from my university account!). But if you do have access, let me know what you think!
https://open.spotify.com/episode/4adN2gVRkQctA55Q0xswiO
#Neuroscience #History #AI #Deepnets #BiologicalIntelligence #BiologicalVision #HumanVision #MachineVision #TheEconomist #Babbage #MachineLearning
hahaha... true. I am too lazy.
if it might cheer you up, even Tony's proof is wrong 😜​
if for all x in R (real), x > 0 AND y/x > -1, then
log (x+y) = log (x) + log (1+y/x)
and the proof from Tony is correct!
So, in Poincare's terms (continuous everywhere, differentiable nowhere), consciousness is a "monster" and an "outrage against human sense", or according to Hermite, a "lamentable scourge"? If so, full agree.
