20 Jan 2024
Our model's connections to hippocampal neurons and AI neural networks has been published in Nature Communications. We uncover signatures predicted by our model in recordings of mouse hippocampus. Inspired by our biological findings, we propose a new loss function that improves neural network performance in multitask learning.
01 Dec 2023
Our model for storing both example-like and concept-like memories has been published in Physical Review E. We present detailed mathematical derivations of its capacities in the mean-field limit as well as simulations that characterize its capabilities to link memories across types.
01 Nov 2023
Louis will be sharing his work with Taro Toyoizumi at Neuroscience 2023. It represents the culmination of a large project on storing memories at both more specific and more generalized scales. Look out soon for our manuscripts, which have already been accepted for publication.
27 Sep 2023
Martin Bjerke will be arriving from the Norwegian University of Science and Technology to work on topological data analysis and other methods for understanding neuroscience data. Welcome!
27 Sep 2023
If you didn't catch Sebastian at Dynamic Days Europe or IBRO, find him today at the Bernstein Conference. He will present his poster “Metabolic dynamics in systems of quadratic integrate and fire neurons and the connection to epilepsy.”
14 Jul 2023
Congratulations to Nacho, who was awarded a prestigious Doctoral INPhINIT fellowship by the ”la Caixa” Foundation to fund his PhD! Thirty fellows across all academic disciplines in Spain are selected each year.
14 Jul 2023
Sebastian will be presenting new work on seizure initiation at the Japan Neuroscience Society conference, which takes place 1–4 August in Sendai. Check out his poster “Initiating seizures in attractor networks: Routes to seizure activity.”
24 Apr 2023
Hannah, an undergraduate at Rensselaer Polytechnic Institute, has joined the Unit for the spring and summer semester. Ignacio, a PhD student at the Complutense University of Madrid, will arrive soon as a CBS Summer Program Intern. Looks like we are in for a vibrant summer.
03 Mar 2023
If you're attending Cosyne 2023, check out “Seeking universality while celebrating heterogeneity among biological attractor networks,” a two-day workshop organized by Louis and John Widloski at UC Berkeley. We will bring together theorists and experimentalists across navigation, working memory, decision making, sensorimotor control, olfaction, machine learning, and other fields. Hope to see you there!
21 Feb 2023
We have posted two preprints about how the hippocampus can retrieve both example-like and concept-like memories. The first comprehensively explores this topic through biological modeling, experimental data analysis, and machine learning. The second derives mathematical formulas that underlie the behavior of our model.
11 Nov 2022
Congraulations to Raymond on publishing his first paper on noise resilience and bump number in continuous attractor networks. Ulya has also published her PhD work on the role of BDNF in the pathogenesis of schizophrenia. These Kang Group interns are making waves.
15 Sep 2022
Ismaeel arrives in Japan from the UK, where he grew up and obtained his PhD in computational chemistry at the University of Manchester. His experience in machine learning will be a great asset to the group as he looks to ply his trade in the field of neuroscience. Welcome!
15 Aug 2022
Raymond is off to do great things at MIT as a PhD student in the Department of Brain and Cognitive Sciences. He finished his undergraduate program at UC Berkeley with an outstanding academic record and a paper accepted for publication in PLOS Computational Biology. We look forward to following his progress and reuniting at conferences.
04 Aug 2022
The Unit goes to Sebastian's hometown, Berlin! We will attend the Bernstein Conference during 13-16 September. Sebastian will present on neural oscillators, and Louis will share his work on memory encodings.
16 Jun 2022
Ulker Isayeva, a PhD student at the University of Cagliari, has been accepted to IBRO-RIKEN CBS Summer Program 2022. We look forward to her internship at the Unit over the next few months, during which she will train RNNs in order to study seizures.
04 Mar 2022
Are you interested in applying machine learning to study seizures and living in Tokyo? We are looking to hire a second talented postdoc or research scientist! View our posting on the RIKEN Center for Brain Science website for more information.
25 Feb 2022
Our bioRxiv preprint “Multiple bumps can enhance robustness to noise in continuous attractor networks” seeks to enhance our fundamental understanding of continuous attractor networks. Simulation results were obtained through the tenacious efforts of Raymond Wang.
07 Feb 2022
Raymond and Louis are excited to present “Multiple bumps can enhance robustness to noise in continuous attractor networks” and “Multiscale encodings of memories in hippocampal and artificial networks” at Cosyne 2022. These projects have been updated with many new results.
20 Sep 2021
Sebastian Eydam will be presenting his recent work on dynamical systems at the Bernstein Conference. Registration is free! His poster is titled “Stochastic excitable system with slowly adapting feedback.” You can catch him on 23 September 2021 at 14:15 CEST.
01 Jul 2021
Welcome to Sebastian, a new postdoc in the group! He arrives from the Technical University of Berlin with a strong quantitative background in physics and dynamical systems. He will be applying his knowledge about coupled networks and phase transitions to study seizures in hippocampal networks.
09 Jun 2021
Check out our new Instagram account! It highlights our engagement with the scientific and public communities that make our research possible and hopefully benefit from our work. It also gives us a way to support similar efforts by others around the world.
09 Apr 2021
Louis joins the Graduate School of Informatics at Kyoto University as an Adjunct Associate Professor. He is delighted to be part of this esteemed institution and looks forward to interacting with students, staff, and faculty.
08 Apr 2021
Our paper “Evaluating state space discovery by persistent cohomology in the spatial representation system” is now published in Frontiers in Computational Neuroscience! We have improved upon our preprint with more rigorous analyses that have produced more impactful conclusions.
11 Feb 2021
If you missed him at the SfN Global Connectome, Raymond will be presenting his work “Multiple bumps enhance robustness to noise in continuous attractor networks” at Cosyne. Visit him virtually at 19:00 US Pacific Time on 26 February 2021.
11 Jan 2021
Are you interested in computational neuroscience, machine learning, and living in Tokyo? The Neural Circuits and Computations Unit is looking to hire a postdoc or research scientist! View our posting on the RIKEN Center for Brain Science website for more information.
08 Dec 2020
Raymond will be presenting his work “Multiple bumps enhance robustness to noise in continuous attractor networks” at the Society for Neuroscience Global Connectome. Visit him virtually at 11:00 am US Pacific Time on 11 January 2021.
01 Nov 2020
Raymond, an undergraduate from University of California, Berkeley, officially joins the group as an intern. He has been working with Louis since March. He is studying how noise in continuous attractor networks produce errors in path integration and methods of suppressing these errors. Welcome!
08 Oct 2020
Read our preprint “State space discovery in spatial representation circuits with persistent cohomology” on bioRxiv. This project explores how topological data analysis can be applied to neuroscience data. It is a collaboration with computational topologists Dmitriy Morozov and Boyan Xu at Lawrence Berkeley National Laboratory.
01 Aug 2020
Welcome to the Neural Circuits and Computations Unit! Louis is excited to start his own research group in computational neuroscience at RIKEN Center for Brain Science. More to come soon.