Biology subjects

Sibindi, T.

Publications and source records attributed to Sibindi, T..

5 recordsLinked to original sources

bioRxiv2020

Lightweight Reinforcement Algorithms for autonomous, scalable intra-cortical Brain Machine Interfaces

Intra-cortical Brain Machine Interfaces (iBMIs) with wireless capability could scale the number of recording channels by integrating an intention decoder to reduce data rates. However, the need for frequent retraining due to neural signal non-stationarity is a big impediment. This paper presents an alternate paradigm of online reinforcement learning (RL) with a binary evaluative feedback in iBMIs to tackle this issue. This paradigm eliminates time-consuming calibration procedures. Instead, it relies on updating the model on a sequential sample-by-sample basis based on an instantaneous evaluative binary feedback signal. However, batch updates of weight in popular deep networks is very resource consuming and incompatible with constraints of an implant. In this work, using offline open-loop analysis on pre-recorded data, we show application of a simple RL algorithm - Banditron -in discrete-state iBMIs and compare it against previously reported state of the art RL algorithms - Hebbian RL, Attention gated RL, deep Q-learning. Owing to its simplistic single-layer architecture, Banditron is found to yield at least two orders of magnitude of reduction in power dissipation compared to state of the art RL algorithms. At the same time, post-hoc analysis performed on four pre-recorded experimental datasets procured from the motor cortex of two non-human primates performing joystick-based movement-related tasks indicate Banditron performing significantly better than state of the art RL algorithms by at least 5%, 10%, 7% and 7% in experiments 1, 2, 3 and 4 respectively. Furthermore, we propose a non-linear variant of Banditron, Banditron-RP, which gives an average improvement of 6%, 2% in decoding accuracy in experiments 2,4 respectively with only a moderate increase in power consumption.

bioengineering↗

bioRxiv2020

Towards Autonomous Intra-cortical Brain Machine Interfaces: Applying Bandit Algorithms for Online Reinforcement Learning

This paper presents application of Banditron - an online reinforcement learning algorithm (RL) in a discrete state intra-cortical Brain Machine Interface (iBMI) setting. We have analyzed two datasets from non-human primates (NHPs) - NHP A and NHP B each performing a 4-option discrete control task over a total of 8 days. Results show average improvements of {approx} 15%, 6% in NHP A and 15%, 21% in NHP B over state of the art algorithms - Hebbian Reinforcement Learning (HRL) and Attention Gated Reinforcement Learning (AGREL) respectively. Apart from yielding a superior decoding performance, Banditron is also the most computationally friendly as it requires two orders of magnitude less multiply-and-accumulate operations than HRL and AGREL. Furthermore, Banditron provides average improvements of at least 40%, 15% in NHPs A, B respectively compared to popularly employed supervised methods - LDA, SVM across test days. These results pave the way towards an alternate paradigm of temporally robust hardware friendly reinforcement learning based iBMIs.

bioengineering↗

bioRxiv2019

Sparse Ensemble Machine Learning to improve robustness of long-term decoding in iBMIs

This paper presents a novel sparse ensemble based machine learning approach to enhance robustness of intracortical Brain Machine Interfaces (iBMIs) in the face of non-stationary distribution of input neural data across time. Each classifier in the ensemble is trained on a randomly sampled (with replacement) set of input channels. These sparse connections ensure that with a high chance, few of the base classifiers should be less affected by the variations in some of the recording channels. We have tested the generality of this technique on different base classifiers - linear discriminant analysis (LDA), support vector machine (SVM), extreme learning machine (ELM) and multilayer perceptron (MLP). Results show decoding accuracy improvements of up to {approx} 21%, 13%, 19%, 10% in non-human primate (NHP) A and 7%, 9%, 7%, 9% in NHP B across test days while using the sparse ensemble approach over a single classifier model for LDA, SVM, ELM and MLP algorithms respectively. The technique also holds ground when the most informative electrode on the test day is dropped. Accordingly, improvements of up to {approx} 24%, 11%, 22%, 9% in NHP A and 14%, 19%, 7%, 28% in NHP B are obtained for LDA, SVM, ELM and MLP respectively.

bioengineering↗

bioRxiv2019

Towards Intelligent Intra-cortical BMI (i2BMI):Low-power Neuromorphic Decoders that outperform Kalman Filters

Fully implantable wireless intra-cortical Brain Machine Interfaces (iBMI) is one of the most promising next frontiers in the nascent field of neurotechnology. However, scaling the number of channels in such systems by another 10X is difficult due to power and bandwidth requirements of the wireless transmitter. One promising solution for that is to include more processing, up to the decoder, in the implant so that transmission data rate is reduced drastically. Earlier work on neuromorphic decoders only showed classification of discrete states. We present results for continuous state decoding using a low power neuromorphic decoder chip termed Spike-input Extreme Learning Machine (SELMA). We compared SELMA against state of the art Steady State Kalman Filter (SSKF) across two different datasets involving a total of 4 non-human primates (NHPs). Results show at least a 10% or more increase in the fraction of variance accounted for by SELMA over SSKF across the datasets. Furthermore, estimated energy consumption comparison shows SELMA consuming {approx} 9 nJ/update against SSKFs {approx} 7.4 nJ/update for an iBMI with a 10 degree of freedom control. Thus, SELMA yields better performance against SSKF with a marginal increase in energy consumption paving the way for reducing transmission data rates in future scaled BMI systems.

bioengineering↗

bioRxiv2019

A neuroanatomically grounded optimal control model of the compensatory eye movement system

We present a working model of the compensatory eye movement system. We challenge the model with a data set of eye movements in mice (n=34) recorded in 4 different sinusoidal stimulus conditions with 36 different combinations of frequency (0.1-3.2 Hz) and amplitude (0.5-8{degrees}) in each condition. The conditions included vestibular stimulation in the dark (vestibular-ocular reflex, VOR), optokinetic stimulation (optokinetic reflex, OKR), and two combined visual/vestibular conditions (the visual-vestibular ocular reflex, vVOR, and visual suppression of the VOR, sVOR). The model successfully reproduced the eye movements in all conditions, except for minor failures to predict phase when gain was very low. Most importantly, it could explain the non-linear summation of VOR and OKR when the two reflexes are activated simultaneously during vVOR stimulation. In addition to our own data, we also reproduced the behavior of the compensatory eye movement system found in the existing literature. These include its response to sum-of-sines stimuli, its response after lesions of the nucleus prepositus hypoglossi or the flocculus, characteristics of VOR adaptation, and characteristics of drift in the dark. Our model is based on ideas of state prediction and forward modeling that have been widely used in the study of motor control. However, it represents one of the first quantitative efforts to simulate the full range of behaviors of a specific system. The model has two separate processing loops, one for vestibular stimulation and one for visual stimulation. Importantly, state prediction in the visual processing loop depends on a forward model of residual retinal slip after vestibular processing. In addition, we hypothesize that adaptation in the system is primarily adaptation of this model. In other words, VOR adaptation happens primarily in the OKR loop.

neuroscience↗