ABC Seminar: Neural networks and neuronal-glial interactions as information processing systems

Welcome to our ABC Seminars! This seminar series is open for everyone. The talks will take place in Otakaari 1, Room A215 (A2). After the talks, coffee and pulla will be served.

The event will be also streamed via Zoom at: https://aalto.zoom.us/j/61448652144

Talks: 

Speaker: Simo Vanni (Helsinki University) 

Title: Neural networks as information processing system, a neuroscience perspective

Abstract: 

Neuroscience has provided a lot of detailed data at multiple levels, but only few studies consider information representation and transmission in neural systems. In contrast, machine learning have information at the core of the field, including theory for learning and representing new information and associated mathematical formalism. It is apparent that neuroscience could benefit from collaboration with the machine learning community, when we start building understanding of the brain as an information processing system. Our recent work examines a recent model of learning representations with a spiking network. Our results are in line with a strong association of information transmission with firing rate. However, a very low firing rate is enough for classifying input to correct output channel. The efficiency of classification led us to propose a refinement to the contemporary theory of predictive coding.

Bio:

Hans Simo Vanni is a specialist in Neurology who holds multiple academic and medical degrees. He is currently a Principal Investigator at the University of Helsinki's Department of Physiology (70% FTE) and a practitioner of Neurology at Mehiläinen Ltd. (30% FTE). He has previously held academic and clinical positions at the University of Helsinki, Aalto University, and Helsinki University Hospital. Vanni has also received funding and grants from various sources, including the European Commission and the Academy of Finland. In addition, he has developed software related to multifocal fMRI and cerebral cortex simulation, and has supervised several doctoral dissertations. Vanni's Orcid identifier is 0000-0002-6101-4129 and his ResearcherID is B-9590-2013. 

Speaker: Marja-Leena Linne (Tampere University)

Title: Neuron-glia interactions: Are they important for brain information processing?

Abstract: 

In my talk I will first briefly introduce state-of-the-art in neuron-glia interactions in mammalian brain. Special emphasis will be on astrocytes which have recently been shown to modulate synaptic transmission, plasticity, and memory as well as take part in many cognitive functions. Using these data and a somatosensory cortical layer 4 to layer 2/3 synapse model developed by us, we have shown that a signal from a postsynaptic neuron, orchestrated by endocannabinoids, astrocytic calcium signaling, and presynaptic N-methyl-D-aspartate receptors coupled with presynaptic calcineurin signaling, can induce timing-dependent long-term depression (t-LTD) which is sensitive to the temporal difference between post- and presynaptic firing. Our results suggest that astrocytes play a critical role in synaptic computation during postnatal development. Towards the end of the talk, I will discuss the possibility that neuron-astrocyte interactions are an important element in rodent and human information processing.

Bio:

Docent Marja-Leena Linne is an engineer and neuroscientist with over 25 years of experience in electrophysiology and computational modeling of neural and brain signals. She is leading Computational Neuroscience Group at Tampere University where she works in the Faculty of Medicine and Health Technology (Group: https://research.tuni.fi/computational-neuroscience/). Her previous position as Academy Research Fellow appointed by the Academy of Finland was at Tampere University of Technology in the Department of Signal Processing. The focus of Dr. Linne’s research is on deciphering the roles of neuronal and glial cells in neural circuits as well as in brain signals recorded from humans in health and disease. Dr. Linne is a core member of EU FET Flagship Human Brain Project (HBP) where she worked as Vice Leader of theoretical neuroscience division to empower innovations in Europe in the intersection of neuroscience, computing, and engineering