AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |
Back to Blog
Dendrite stories3/2/2024 ![]() Functional clustering of dendritic activity during decision-making. Dendritic spine geometry is critical for AMPA receptor expression in hippocampal CA1 pyramidal neurons. Dendritic discrimination of temporal input sequences in cortical neurons. NMDA receptors: linking physiological output to biophysical operation. Signal delay and input synchronization in passive dendritic structures. Neighbouring spines are activated serially along a dendrite, towards or away from the cell body.Īgmonsnir, H. Locally sequential synaptic reactivation during hippocampal ripples. Reliable sequential activation of neural assemblies by single pyramidal cells in a three-layered cortex. Hemberger, M., Shein-Idelson, M., Pammer, L. Scaling of brain metabolism with a fixed energy budget per neuron: implications for neuronal activity, plasticity and evolution. In International Conference on Learning Representations (ICLR, 2021). Isotropy in the contextual embedding space: Clusters and manifolds. In 33rd Conference on Neural Information Processing Systems (NeurIPS, 2019)Ĭai, X., Huang, J., Bian, Y. Deep ReLU networks have surprisingly few activation patterns. A specific sequence of spikes encodes memory of an episode in humans and recall involves reinstating this temporal order of activity. Replay of cortical spiking sequences during human memory retrieval. Odour encoding by temporal sequences of firing in oscillating neural assemblies. Replay of neuronal firing sequences in rat hippocampus during sleep following spatial experience. Spike-based strategies for rapid processing. 3D-aCortex: an ultra-compact energy-efficient neurocomputing platform based on commercial 3D-NAND flash memories. 3-D stacked synapse array based on charge-trap flash memory for implementation of deep neural networks. Reconstruction and simulation of neocortical microcircuitry. In 2022 IEEE International Solid-State Circuits Conference (ISSCC) 1–3 (IEEE, 2022). A 1-Tb Density 4b/Cell 3D-NAND Flash on 176-Tier Technology with 4-Independent Planes for Read using CMOS-Under-the-Array. 3-D NAND technology achievements and future scaling perspectives. In 2018 IEEE Symposium on VLSI Circuits 3–6 (IEEE, 2018). Hardware-enabled artificial intelligence. Carbontracker: tracking and predicting the carbon footprint of training deep learning models. Language models are unsupervised multitask learners. Deep learning in neural networks: an overview. This paper introduced the synaptocentric conception of the learning brain. ![]() The perceptron-a probabilistic model for information-storage and organization in the brain. A domain-specific supercomputer for training deep neural networks. Brain-inspired computing needs a master plan. Are you scared yet, human? The Guardian (2020). With the help of a computational model of a dendrite and a conceptual model of a ferroelectric device that emulates it, I illustrate how dendrocentric learning artificial intelligence-or synthetic intelligence for short-could run not with megawatts in the cloud but rather with watts on a smartphone.īrown, T. Synaptic inputs are not weighted precisely but rather ordered meticulously along a short stretch of dendrite, termed dendrocentric learning. Here I propose to transcend this three-dimensional thermal constraint by moving away from learning with synapses to learning with dendrites. Although travel can be shortened by stacking tiled multipliers in a three-dimensional chip, such a solution acutely reduces the available surface area for dissipating heat. ![]() Moreover, the returns from tiling these multipliers ever more densely now diminish because signals must travel relatively farther and farther. Artificial intelligence now advances by performing twice as many floating-point multiplications every two months, but the semiconductor industry tiles twice as many multipliers on a chip every two years.
0 Comments
Read More
Leave a Reply. |