Dr. Robert A. Vandermeulen presents The 54th edition of the PRAGUE COMPUTER SCIENCE SEMINAR
On 2023-04-20 16:15:00 at KN:E-107, Karlovo náměstí 13, Praha 2
Inferring and Comparing the Concept Spaces of Humans and Neural Networks
The lecture will be followed by a discussion
One fundamental goal of cognitive science is to understand how humans mentally
represent concepts. One approach to this is to survey humans how they perceive
objects and their similarity. The first part of this talk will cover VICE, a
Bayesian method for inferring geometric representations of concepts, i.e. vector
embeddings, whose geometry is consistent with similarity task responses.
Representations from VICE are very effective at predicting human responses to
object similarity tasks. Representation dimensions from VICE are highly
interpretable, thus revealing what properties are important when humans consider
object similarity. The second part of this talk will investigate how these human
representations compare to neural network representations on the THINGS dataset,
a collection of 1,854 object categories carefully curated for cognitive science
research. Here we see what kinds of networks are best aligned to human
understanding and what object properties that humans find important are also
understood by neural networks.
Dr. Robert A. Vandermeulen received his PhD in ElectricalEngineering from the
University of Michigan in 2016. He was a postdoctoral researcher under Prof. Dr.
Marius Kloft at Technical University Kaiserslautern from 2017 to 2020 and is
currently a senior researcher at the Berlin Institute for the Foundations of
Learning and Data (BIFOLD). Dr. Vandermeulen’s work spans many topics, however
his most prominent work concerns deep anomaly detection where his work on deep
one-class classification was awarded the SIGKDD Workshop on Anomaly and Novelty
Detection, Explanation and Accommodation Test of Time Award.
The lecture will be followed by a discussion
One fundamental goal of cognitive science is to understand how humans mentally
represent concepts. One approach to this is to survey humans how they perceive
objects and their similarity. The first part of this talk will cover VICE, a
Bayesian method for inferring geometric representations of concepts, i.e. vector
embeddings, whose geometry is consistent with similarity task responses.
Representations from VICE are very effective at predicting human responses to
object similarity tasks. Representation dimensions from VICE are highly
interpretable, thus revealing what properties are important when humans consider
object similarity. The second part of this talk will investigate how these human
representations compare to neural network representations on the THINGS dataset,
a collection of 1,854 object categories carefully curated for cognitive science
research. Here we see what kinds of networks are best aligned to human
understanding and what object properties that humans find important are also
understood by neural networks.
Dr. Robert A. Vandermeulen received his PhD in ElectricalEngineering from the
University of Michigan in 2016. He was a postdoctoral researcher under Prof. Dr.
Marius Kloft at Technical University Kaiserslautern from 2017 to 2020 and is
currently a senior researcher at the Berlin Institute for the Foundations of
Learning and Data (BIFOLD). Dr. Vandermeulen’s work spans many topics, however
his most prominent work concerns deep anomaly detection where his work on deep
one-class classification was awarded the SIGKDD Workshop on Anomaly and Novelty
Detection, Explanation and Accommodation Test of Time Award.
External www: https://www.praguecomputerscience.cz/