This special online lecture series will take place twice a month on Wednesdays from 10:00 am to 11:00 am EST (4:00 pm – 5:00pm CET). An unpublished work in progress serves as a basis for this lecture series. A total of six to eight lectures are being planned. Register here!
Each lecture will be 30 minutes long and followed by 30 minutes of Q&A. One week later, a hands-on session will apply knowledge; please bring your own case data! Your favorite programming language can be used (Matlab, Python, R). The code is available in Matlab and in Python (use Python3 with numpy, scipy, and matplotlib). The associated code can be found here.
Wednesdays at 10:00 am – 11:00 am EST (4:00 pm – 5:00pm CET)
April 13th 2022 (30 min lecture + 30 min questions)
Fundamental distributions and Bayesian parameters
Missed the lecture? Enjoy it here.
April 20th 2022 (1h hands-on)
Putting Bayesian parameters to work
Missed the lecture? Enjoy it here.
May 4th 2022 (30 min lecture + 30 min questions)
Fundamentals of Information Theory
Missed the lecture? Enjoy it here.
Entropy of a continuous distribution.
Available information
Shannon’s Coding Theorem
The Kullback-Leibler Divergence
Temporal Information
The Time Scale Invariance of Temporal Information
Communicated Information
Contingency
May 11th 2022 (1h hands-on)
Applying simple Information Theory to associative learning (hands-on)
Missed the lecture? Enjoy it here.
June 1st 2022
Applications
Missed the lecture? Enjoy it here.
Time-Scale Invariance and Contingency in Reinforcement Learning
Estimating Prospective Contingencies After the First Few Rewards
Estimating Retrospective Contingencies After the First Few Rewards
Measuring The Strength of the Evidence
Applying the nDkl in a Reinforcement Learning Paradigm (aka Operant Conditioning)
Measuring the Strength of the Evidence for Differences in Probability
Checking on the stability of a parameter estimate
Measuring the Growing Strength of Stochastic Stimuli
The Growth of Behavior-Independent Probability Estimates
Tracking the Change in the Behavioral Probabilities
Measuring Contingency Detection Behaviorally and Photometrically
Integrating over the Posterior Distributions
Applying the nDkl to Photometric Data on DA activity.
June 8th 2022
Continue from previous session & Conclusions