2021/05/19 by Jacob Russin, Maryam Zolfaghar, Russin, Jacob +7 · 1 citation
Computer Science · Neuroscience · #Domain Adaptation and Few-Shot Learning #FOS: Biological sciences #FOS: Computer and information sciences #Machine Learning (cs.LG) #Memory and Neural Mechanisms #Neural Networks and Applications #Neuroinflammation and Neurodegeneration Mechanisms #Neurons and Cognition (q-bio.NC) #Neuroscience and Neuropharmacology Research
paper · pdf · doi:10.48550/arxiv.2105.08944
openalex publication_date 2021/05/19 · openalex created_date 2021/08/02 · openalex updated_date 2026/07/28
The neural mechanisms supporting flexible relational inferences, especially\nin novel situations, are a major focus of current research. In the\ncomplementary learning systems framework, pattern separation in the hippocampus\nallows rapid learning in novel environments, while slower learning in neocortex\naccumulates small weight changes to extract systematic structure from\nwell-learned environments. In this work, we adapt this framework to a task from\na recent fMRI experiment where novel transitive inferences must be made\naccording to implicit relational structure. We show that computational models\ncapturing the basic cognitive properties of these two systems can explain\nrelational transitive inferences in both familiar and novel environments, and\nreproduce key phenomena observed in the fMRI experiment.\n