CURRENT
Neural Dynamics of Stuttering
Computational modeling of how neural dynamics support fluent speech motor control, and what goes wrong in stuttering. Details will follow publication.
LINK ESTABLISHED · BOSTON UNIVERSITY · NEURAL ENGINEERING
Ram J.
Zaveri
> ▍
I take biological learning apart to build machines that learn the same way. Today that means biophysically grounded models of neural dynamics and the analysis of electrophysiological and behavioral time series, aimed at neuromimetic systems that help design treatments for neurological disorders.
0years of research
0peer-reviewed papers
0venues reviewed for
0awards & honors
01
Selected work
Intracranial Theta & Navigation
Full pipeline for NeuroPace RNS iEEG, from artifact rejection to theta bouts measured against the aperiodic background and related to behavior.
- Per-channel artifact and epileptiform-transient rejection, detrending, line-noise removal and bandpass filtering within the RNS bandwidth.
- Theta-band (~4–12 Hz) instantaneous amplitude and phase from time-frequency decomposition, with oscillatory bouts quantified against a fitted 1/f background.
- Synchronized neural, behavioral and positional streams. Circular–linear statistics relate theta to distance-to-target and spatial location.
SiamABC: Efficient Robust Tracking
A real-time Siamese tracker that adapts at test time to adverse visibility on CPU-only edge devices.
- Efficient trackers overfit to in-distribution data. SiamABC improves accuracy markedly on out-of-distribution sequences.
- A forward-only, backward-free update rule on batch-norm statistics adapts to the incoming stream at negligible latency.
- CPU-efficient template–search attention and new self-supervised losses offset the limited capacity of efficient trackers.
02
Explore the archive
/about
About
Background, education, the full experience trajectory, honors, peer review and toolkit.
ENTER → /researchResearch
Fifteen projects across neuroscience, signal processing, learning theory, vision and industry, each with a live visual.
ENTER → /publicationsPublications
Peer-reviewed papers at WACV, ECCV-W, ISBI and CVPR-W, plus talks and posters.
ENTER →
03
Open a channel
I'm glad to hear about collaborations on neural dynamics, BCI, neuromorphic computing or biologically inspired learning, or just to talk about brains and machines.
EMAIL zaveri@bu.edu
LOCATION Boston, MA, USA
SCHOLAR Google Scholar ↗
LINKEDIN /in/ramzaveri ↗
GITHUB rz0012 ↗