Abstract
Neuroprosthetic brain-computer interfaces function via an algorithm which decodes neural activity of the user into movements of an end effector, such as a cursor or robotic arm. In practice, the decoder is often learned by updating its parameters while the user performs a task. When the user’s intention is not directly observable, recent methods have demonstrated value in training the decoder against a surrogate for the user’s intended movement. Here we show that training a decoder in this way is a novel variant of an imitation learning problem, where an oracle or expert is employed for supervised training in lieu of direct observations, which are not available. Specifically, we describe how a generic imitation learning meta-algorithm, dataset aggregation (DAgger), can be adapted to train a generic brain-computer interface. By deriving existing learning algorithms for brain-computer interfaces in this framework, we provide a novel analysis of regret (an important metric of learning efficacy) for brain-computer interfaces. This analysis allows us to characterize the space of algorithmic variants and bounds on their regret rates. Existing approaches for decoder learning have been performed in the cursor control setting, but the available design principles for these decoders are such that it has been impossible to scale them to naturalistic settings. Leveraging our findings, we then offer an algorithm that combines imitation learning with optimal control, which should allow for training of arbitrary effectors for which optimal control can generate goal-oriented control. We demonstrate this novel and general BCI algorithm with simulated neuroprosthetic control of a 26 degree-of-freedom model of an arm, a sophisticated and realistic end effector.
Highlights
Brain-computer interfaces (BCI, or brain-machine interfaces) translate noisy neural activity into commands for controlling an effector via a decoding algorithm [1,2,3,4]
We offer a unification of closed-loop decoder learning setting as an imitation learning problem
In the brain computer interface (BCI) setting, instead of a policy that asserts which action to take in a given state, we have a decoder that determines the effector update in response to the current kinematic state and neural activity
Summary
Brain-computer interfaces (BCI, or brain-machine interfaces) translate noisy neural activity into commands for controlling an effector via a decoding algorithm [1,2,3,4]. In many cases of interest the user is not able to make overt movements, so intended movements must be inferred or otherwise determined This insight that better decoder parameters can be learned by training against some form of assumed intention appears in [11], and extensions have been explored in [13, 14]. In these works, it is assumed that the user intends to move towards the current goal or target in a cursor task, resulting in parameter training algorithms that result in dramatically improved decoder performance on a cursor task
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.