no code implementations • 27 Oct 2021 • Guangzhi Tang, Neelesh Kumar, Ioannis Polykretis, Konstantinos P. Michmizos
We propose a biologically plausible gradient-based learning algorithm for SNN that is functionally equivalent to backprop, while adhering to all three neuromorphic principles.
1 code implementation • NeurIPS 2021 • Vladimir A. Ivanov, Konstantinos P. Michmizos
With a top accuracy of 97. 61% on MNIST, 97. 51% on N-MNIST, and 85. 84% on Fashion-MNIST, NALSM achieved comparable performance to current fully-connected multi-layer spiking neural networks trained via backpropagation.
1 code implementation • 19 Oct 2020 • Guangzhi Tang, Neelesh Kumar, Raymond Yoo, Konstantinos P. Michmizos
Here, we propose a population-coded spiking actor network (PopSAN) trained in conjunction with a deep critic network using deep reinforcement learning (DRL).
no code implementations • 8 Jun 2020 • Ioannis Polykretis, Konstantinos P. Michmizos
Locomotion is a crucial challenge for legged robots that is addressed "effortlessly" by biological networks abundant in nature, named central pattern generators (CPG).
1 code implementation • 2 Mar 2020 • Guangzhi Tang, Neelesh Kumar, Konstantinos P. Michmizos
Here, we propose a neuromorphic approach that combines the energy-efficiency of spiking neural networks with the optimality of DRL and benchmark it in learning control policies for mapless navigation.
no code implementations • 18 Feb 2020 • Neelesh Kumar, Konstantinos P. Michmizos
Although cognitive engagement (CE) is crucial for motor learning, it remains underutilized in rehabilitation robots, partly because its assessment currently relies on subjective and gross measurements taken intermittently.
no code implementations • 18 Feb 2020 • Praveenram Balachandar, Konstantinos P. Michmizos
Robotic vision introduces requirements for real-time processing of fast-varying, noisy information in a continuously changing environment.
no code implementations • 18 Feb 2020 • Neelesh Kumar, Konstantinos P. Michmizos
Here, we propose a deep convolutional neural network (CNN) that uses electroencephalography (EEG) as an objective measurement of two kinematics components that are typically used to assess motor learning and thereby adaptation: i) the intent to initiate a goal-directed movement, and ii) the reaction time (RT) of that movement.
1 code implementation • 2 Jul 2019 • Guangzhi Tang, Ioannis E. Polykretis, Vladimir A. Ivanov, Arpit Shah, Konstantinos P. Michmizos
While there is still a lot to learn about astrocytes and their neuromodulatory role in the spatial and temporal integration of neuronal activity, their introduction to neuromorphic hardware is timely, facilitating their computational exploration in basic science questions as well as their exploitation in real-world applications.
no code implementations • 22 Mar 2019 • Vladimir A. Ivanov, Ioannis E. Polykretis, Konstantinos P. Michmizos
Increasing experimental evidence suggests that axonal action potential conduction velocity is a highly adaptive parameter in the adult central nervous system.
no code implementations • 18 Mar 2019 • Ioannis E. Polykretis, Vladimir A. Ivanov, Konstantinos P. Michmizos
Deciphering the complex interactions between neurotransmission and astrocytic $Ca^{2+}$ elevations is a target promising a comprehensive understanding of brain function.
no code implementations • 6 Mar 2019 • Guangzhi Tang, Arpit Shah, Konstantinos P. Michmizos
We performed comparative analyses for accuracy and energy-efficiency between our neuromorphic approach and the GMapping algorithm, which is widely used in small environments.
no code implementations • 5 Jul 2018 • Guangzhi Tang, Konstantinos P. Michmizos
It is true that the "best" neural network is not necessarily the one with the most "brain-like" behavior.
no code implementations • 13 Feb 2017 • Leo Kozachkov, Konstantinos P. Michmizos
Finding the origin of slow and infra-slow oscillations could reveal or explain brain mechanisms in health and disease.