Elijah A. Taeckens
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Elijah A. Taeckens has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Reza Shadmehr
Laboratory for Computational Motor Control, Department of Biomedical Engineering, Johns Hopkins School of Medicine, Baltimore MD
2 shared publications - Alana D. Chandler
Laboratory for Computational Motor Control, Department of Biomedical Engineering, Johns Hopkins School of Medicine, Baltimore MD
1 shared publication - Alden M. Shoup
Johns Hopkins School of Medicine
1 shared publication - Alexander Nicholas Vasserman
Laboratory for Computational Motor Control, Department of Biomedical Engineering, Johns Hopkins School of Medicine, Baltimore MD
1 shared publication - Hisham Y. Elseweifi
Laboratory for Computational Motor Control, Department of Biomedical Engineering, Johns Hopkins School of Medicine, Baltimore MD
1 shared publication - Jafar Doostmohammadi
Laboratory for Computational Motor Control, Department of Biomedical Engineering, Johns Hopkins School of Medicine, Baltimore MD
1 shared publication
Latest Publications
- Open Access
Climbing fibers encode the gradient of a loss function for the cerebellum
Neurons in the brain are often many synapses away from motoneurons, yet if a movement results in error, each distant neuron needs a teacher that considers its specific contribution to production of that movement. This credit assignment problem is solved in machine learning via gradient descent of a loss function, where the loss defines the subjective cost incurred by error. Does the brain use gradient descent to teach individual neurons? We trained marmosets to make saccades to visual targets and then varied the loss by assigning reward value to each target. The climbing fibers, which are the teachers of Purkinje cells (P-cells) in the cerebellum, used a multiplicative encoding to scale the spatial properties of the error vector with its reward properties, incorporating reward prediction errors. Using spike-triggered suppression of P-cells, we quantified the potent vector that mapped each P-cell’s output to eye movements and discovered that the climbing fibers were not merely transmitting errors. Rather, they were providing a signal that was, on average, proportional to the dot product of the reward dependent error vector upon the P-cell’s potent vector. Thus, the climbing fibers solved the credit assignment problem by providing the gradient of a loss function with respect to the output of their individual P-cells.
Open AccessClimbing fibers encode the gradient of a loss function for the cerebellum
Neurons in the brain are often many synapses away from motoneurons, yet if a movement results in error, each distant neuron needs a teacher that considers its specific contribution to production of that movement. This credit assignment problem is solved in machine learning via gradient descent of a loss function, where the loss defines the subjective cost incurred by error. Does the brain use gradient descent to teach individual neurons? We trained marmosets to make saccades to visual targets and then varied the loss by assigning reward value to each target. The climbing fibers, which are the teachers of Purkinje cells (P-cells) in the cerebellum, used a multiplicative encoding to scale the spatial properties of the error vector with its reward properties, incorporating reward prediction errors. Using spike-triggered suppression of P-cells, we quantified the potent vector that mapped each P-cell’s output to eye movements and discovered that the climbing fibers were not merely transmitting errors. Rather, they were providing a signal that was, on average, proportional to the dot product of the reward dependent error vector upon the P-cell’s potent vector. Thus, the climbing fibers solved the credit assignment problem by providing the gradient of a loss function with respect to the output of their individual P-cells.
- Open Access
Coordination of spike timing among the neurons of the cerebellum
We tend to think of neurons as either excitatory or inhibitory, but certain neurons chemically inhibit their downstream targets while electrically exciting each other. For example, in the cerebellum, molecular layer interneurons type 1 (MLI1s) inhibit Purkinje cells (P-cells) via release of GABA but promote spiking in each other via gap junctions. P-cells inhibit nucleus neurons while exciting each other via ephaptic coupling. What is gained by excitatory interactions among inhibitory neurons? We recorded from the marmoset cerebellum during saccadic eye movements and found that spike timing in electrically coupled P-cell pairs, as well as MLI1 pairs, exhibited a mathematical regularity: as firing rates increased, the rate of spikes that were within 1ms of each other grew disproportionately while 2-4ms intervals were suppressed. We isolated triplets in which two MLI1s converged onto a single P-cell and found that if the MLI1s spiked within 1ms of each other, they produced superposition of their individual effects on their target; a deep inhibition followed by a post-inhibitory rebound. This enhanced the temporal precision in the downstream P-cell’s next spike. However, when the MLI1s spiked 2-4ms apart, the two spikes interfered with each other, producing partial cancellation. Thus, electrical coupling of inhibitory neurons promoted production of spike intervals that induced constructive superposition. This reduced the variance of spike timing in the downstream neuron.
Open AccessCoordination of spike timing among the neurons of the cerebellum
We tend to think of neurons as either excitatory or inhibitory, but certain neurons chemically inhibit their downstream targets while electrically exciting each other. For example, in the cerebellum, molecular layer interneurons type 1 (MLI1s) inhibit Purkinje cells (P-cells) via release of GABA but promote spiking in each other via gap junctions. P-cells inhibit nucleus neurons while exciting each other via ephaptic coupling. What is gained by excitatory interactions among inhibitory neurons? We recorded from the marmoset cerebellum during saccadic eye movements and found that spike timing in electrically coupled P-cell pairs, as well as MLI1 pairs, exhibited a mathematical regularity: as firing rates increased, the rate of spikes that were within 1ms of each other grew disproportionately while 2-4ms intervals were suppressed. We isolated triplets in which two MLI1s converged onto a single P-cell and found that if the MLI1s spiked within 1ms of each other, they produced superposition of their individual effects on their target; a deep inhibition followed by a post-inhibitory rebound. This enhanced the temporal precision in the downstream P-cell’s next spike. However, when the MLI1s spiked 2-4ms apart, the two spikes interfered with each other, producing partial cancellation. Thus, electrical coupling of inhibitory neurons promoted production of spike intervals that induced constructive superposition. This reduced the variance of spike timing in the downstream neuron.
- Open Access
- 1 December 2023
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 20 (6) , 066042
A spiking neural network with continuous local learning for robust online brain machine interface
Objective. Spiking neural networks (SNNs) are powerful tools that are well suited for brain machine interfaces (BMI) due to their similarity to biological neural systems and computational efficiency. They have shown comparable accuracy to state-of-the-art methods, but current training methods require large amounts of memory, and they cannot be trained on a continuous input stream without pausing periodically to perform backpropagation. An ideal BMI should be capable training continuously without interruption to minimize disruption to the user and adapt to changing neural environments. Approach. We propose a continuous SNN weight update algorithm that can be trained to perform regression learning with no need for storing past spiking events in memory. As a result, the amount of memory needed for training is constant regardless of the input duration. We evaluate the accuracy of the network on recordings of neural data taken from the premotor cortex of a primate performing reaching tasks. Additionally, we evaluate the SNN in a simulated closed loop environment and observe its ability to adapt to sudden changes in the input neural structure. Main results. The continuous learning SNN achieves the same peak correlation ( ρ = 0.7 ) as existing SNN training methods when trained offline on real neural data while reducing the total memory usage by 92%. Additionally, it matches state-of-the-art accuracy in a closed loop environment, demonstrates adaptability when subjected to multiple types of neural input disruptions, and is capable of being trained online without any prior offline training. Significance. This work presents a neural decoding algorithm that can be trained rapidly in a closed loop setting. The algorithm increases the speed of acclimating a new user to the system and also can adapt to sudden changes in neural behavior with minimal disruption to the user.
- 1 December 2023
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 20 (6) , 066042
Open AccessA spiking neural network with continuous local learning for robust online brain machine interface
Objective. Spiking neural networks (SNNs) are powerful tools that are well suited for brain machine interfaces (BMI) due to their similarity to biological neural systems and computational efficiency. They have shown comparable accuracy to state-of-the-art methods, but current training methods require large amounts of memory, and they cannot be trained on a continuous input stream without pausing periodically to perform backpropagation. An ideal BMI should be capable training continuously without interruption to minimize disruption to the user and adapt to changing neural environments. Approach. We propose a continuous SNN weight update algorithm that can be trained to perform regression learning with no need for storing past spiking events in memory. As a result, the amount of memory needed for training is constant regardless of the input duration. We evaluate the accuracy of the network on recordings of neural data taken from the premotor cortex of a primate performing reaching tasks. Additionally, we evaluate the SNN in a simulated closed loop environment and observe its ability to adapt to sudden changes in the input neural structure. Main results. The continuous learning SNN achieves the same peak correlation ( ρ = 0.7 ) as existing SNN training methods when trained offline on real neural data while reducing the total memory usage by 92%. Additionally, it matches state-of-the-art accuracy in a closed loop environment, demonstrates adaptability when subjected to multiple types of neural input disruptions, and is capable of being trained online without any prior offline training. Significance. This work presents a neural decoding algorithm that can be trained rapidly in a closed loop setting. The algorithm increases the speed of acclimating a new user to the system and also can adapt to sudden changes in neural behavior with minimal disruption to the user.
- Open Access
A Spiking Neural Network with Continuous Local Learning for Robust Online Brain Machine Interface
Objective: Spiking neural networks (SNNs) are powerful tools that are well suited for brain machine interfaces (BMI) due to their similarity to biological neural systems and computational efficiency. They have shown comparable accuracy to state-of-the-art methods, but current training methods require large amounts of memory, and they cannot be trained on a continuous input stream without pausing periodically to perform backpropagation. An ideal BMI should be capable training continuously without interruption to minimize disruption to the user and adapt to changing neural environments. Approach: We propose a continuous SNN weight update algorithm that can be trained to perform regression learning with no need for storing past spiking events in memory. As a result, the amount of memory needed for training is constant regardless of the input duration. We evaluate the accuracy of the network on recordings of neural data taken from the premotor cortex of a primate performing reaching tasks. Additionally, we evaluate the SNN in a simulated closed loop environment and observe its ability to adapt to sudden changes in the input neural structure. Main results: The continuous learning SNN achieves the same peak correlation ( ρ = 0.7) as existing SNN training methods when trained offline on real neural data while reducing the total memory usage by 92%. Additionally, it matches state-of-the-art accuracy in a closed loop environment, demonstrates adaptability when subjected to multiple types of neural input disruptions, and is capable of being trained online without any prior offline training. Significance: This work presents a neural decoding algorithm that can be trained rapidly in a closed loop setting. The algorithm increases the speed of acclimating a new user to the system and also can adapt to sudden changes in neural behavior with minimal disruption to the user.
Open AccessA Spiking Neural Network with Continuous Local Learning for Robust Online Brain Machine Interface
Objective: Spiking neural networks (SNNs) are powerful tools that are well suited for brain machine interfaces (BMI) due to their similarity to biological neural systems and computational efficiency. They have shown comparable accuracy to state-of-the-art methods, but current training methods require large amounts of memory, and they cannot be trained on a continuous input stream without pausing periodically to perform backpropagation. An ideal BMI should be capable training continuously without interruption to minimize disruption to the user and adapt to changing neural environments. Approach: We propose a continuous SNN weight update algorithm that can be trained to perform regression learning with no need for storing past spiking events in memory. As a result, the amount of memory needed for training is constant regardless of the input duration. We evaluate the accuracy of the network on recordings of neural data taken from the premotor cortex of a primate performing reaching tasks. Additionally, we evaluate the SNN in a simulated closed loop environment and observe its ability to adapt to sudden changes in the input neural structure. Main results: The continuous learning SNN achieves the same peak correlation ( ρ = 0.7) as existing SNN training methods when trained offline on real neural data while reducing the total memory usage by 92%. Additionally, it matches state-of-the-art accuracy in a closed loop environment, demonstrates adaptability when subjected to multiple types of neural input disruptions, and is capable of being trained online without any prior offline training. Significance: This work presents a neural decoding algorithm that can be trained rapidly in a closed loop setting. The algorithm increases the speed of acclimating a new user to the system and also can adapt to sudden changes in neural behavior with minimal disruption to the user.