Inferring Visual Function and Organization via a Functional Manifold of Neurons, Steven Zucker

Описание: Different stimuli express different activity patterns in different neurons. As a step toward understanding visual circuitry, it is necessary to organize these differences. We introduce a manifold in which every point is a neuron. Since nearby neurons respond similarly (in time) to similar components of the stimulus ensemble, it is likely that these nearby neurons participate in functional circuits.

This in-person workshop was the culmination of our on-going multidisciplinary exploration project Understanding the Nature of Inference: Correlation and Causation. During the course of our colloquium series, generously funded by the John Templeton Foundation, we explored how inference models operate across disciplines by learning from each other. To this end, we endeavored to go beyond our respective vantage points, across fields and into a new epistemic framework to define causal relationships and how they function. In particular, we discussed the various kinds of methodological schemas, their merits and limits and potential for refinement and re-definition to ferret out causal connections.

During our December 2023 Workshop, experts from varied disciplines presented how they set up problem solving given the complexity of systems that they model; the philosophers assembled examined the nature of laws. A key question they were asked to address in addition to explaining the current landscape of modeling methodologies was how a near-future data deluge is likely to impact their modeling methodologies. Most fields stand to transform dramatically with the influx of vast amounts of new data expected within the next 2 – 5 years. How current conceptual models will need to be refined and altered in this scenario were discussed within the talks and amongst our numerous participants.

We are indebted to The Edward J. and Dorothy Clarke Kempf Memorial Fund and The Whitney and Betty MacMillan Center for International and Area Studies at Yale, as well as to the John Templeton Foundation, for the generous funding we received to bring this Workshop to fruition.