John Pearson

@jmxpearson.bsky.social

1523 Followers 969 Following

Computational neuroscience, neuroML, natural behavior. I charge more for miracles. PI @ pearsonlab.github.io. Center for Neural Science, NYU and Center for Computational Neuroscience, @flatironinstitute.org.

years later. at meeting, held two In November 1958 Uttley (1906-85) organized a select four-day seminar at Lon-don's NPL. He himself had worked for some years at the UK's Radar Research Establishment, but his interests went way beyond the technicalities of radar. He'd done pioneering work in symbolic Al (a logic program, for instance: Uttley 1951), in information-theoretic neurology (1954), and in connectionist modelling (1956). So he was eminently well suited to host a seminar that promised to set many intellectual hares running. And that, it certainly did. What the participants lacked in quantity, they more than made up in quality. The event turned out to be important for both connectionist and symbolic Al, and for computational neuroscience too. Indeed, the list of papers given at this small meeting reads almost like a core course on cognitive science. For among NPL's many memorable moments were these: * Selfridge first presented Pandemonium (Chapter 12.ii.d). * Frank Rosenblatt introduced perceptrons (12.ii.e). * Barlow announced his 'coding' theory of perception (14.iii.b). * Gregory initiated his cat-among-pigeons critique of brain-ablation studies (14.jii.a). * MacKay made public his ideas about building 'hybrid' (analogue-digital) ma-chines, previously discussed only privately —with the Ratio members, for instance (4.v.b and 12.ix.b). * McCarthy proposed that programs could be given "common sense" by expressing everyday knowledge in Russell's predicate calculus, using a programming language later developed as LISP (10.i.f, iii.e, and v.c). * Minsky talked about heuristic programming, and told the soon-to-be-garbled story of the 'computer-generated' geometry proof (10.i.c). This was the first officially published version of core ideas in his 'Steps' paper (10.i.g), already handed out in draft at Dartmouth. * Pask described his electrochemical model of a dynamically developing concept (4.v.e). * And, as a dissenting voice, Yehoshua Bar-Hillel argued that McCar…
Picture of a girl dressed as a cowboy and a baby with a tail pinned on it. “Today I am a cowgirl,” she said to Baby Dodd. “And you will be the cow.”
A diagram of actor-critic reinforcement learning in which the critic calculates a negative reward (error) via a local circuit of the type discussed above.
A learning diagram in which a curved surface (the loss landscape) changes shape and minimum in the first three panels (labeled "Sensory phase"). In the final panel ("Sensorimotor phase"), a dot representing song descends the loss landscape.
On the left, "Phase I: learn landscape modes" shows a dotted red line like a shallow bowl curve up into a vase-shaped solid red line. On the right, "Phase II: learn memory modes" shows the dotted blue vase shape (left) shift rightward to a point labeled "tutor song."
Two scatterplots showing the responses of CM neurons to correct singing (x axis) versus either perturbed singing (top plot) or singing post-deafening (bottom plot). Both plots have most points along the diagonal, with a few outliers above, indicated by red arrows.
Image plots showing activations of a neural population to correct singing, singing under weak perturbation, singing under strong perturbation, and singing post-deafening.
Diagram of a neural network with a premotor population projecting to a local population of excitatory neurons reciprocally connected to a local population of inhibitory neurons.
A diagram of the problem setup: a box labeled sparse coding sends an arrow to one labeled "auditory input," which in turn sends an arrow to a circle labeled "secondary auditory nuclei." Another box labeled "premotor" also sends an arrow to the circle.
Similar to the previous image. Rows of shapes at the top represent the changing structure of the bird's own song. On the row below, a single repeated shape represents the learned structure of the tutor song. Below, line drawings represent responses of two neurons that diminish over the course of learning as the bird's own song replicates the tutor song.
A diagram showing two rows of two shapes each. The upper row is labeled "Tutor's Song," and the lower row is labeled "Early Tutoring" and "Late Tutoring." The shapes in the first column do not match, but the shapes in the second column are mirror images of each other.
A male zebra finch with his beak turned to a juvenile zebra finch.
A juvenile zebra finch.
A picture of a bedraggled cat captioned “i fear i have edited my own soul down to 500 words.”
Plot of posterior distributions of MCMC samples.
A kid and his dog looking skeptical.