REVBRAIN: Real-time Probabilistic EEG Forecasting Using Residual Vector Quantization
Abstract
Real-time forecasting of resting-state EEG signals is required by applications like brain-state-dependent stimulation, where computer generated actions are coupled with ongoing neural oscillations. Recently, the deep learning model WaveNet was proposed for the forecasting task. It shows superior accuracy compared to the prevailing method, the linear autoregressive (AR) model, and provides a built-in uncertainty measure due to its probabilistic modeling framework. However, inference using WaveNet is too slow for real-time use due to its sample-by-sample autoregressive operation: a single forecast requires evaluating the model tens or hundreds of times depending on the length of the prediction window. In this paper, we present REVBRAIN, a probabilistic deep learning method that produces forecasts in real time by generating the forecast over the whole prediction horizon at once. This is achieved by tokenizing the output space using residual vector quantization (RVQ) and generating forecasts over the items in the tokenized space. We tested REVBRAIN for forecasting 200 ms of EEG time series in the alpha-frequency band (8-13 Hz) and compared our results against WaveNet model. We found that REVBRAIN achieved real-time inference on a laptop GPU and an over 50× speedup compared to WaveNet, while reaching superior prediction accuracy. By achieving real-time inference speed, REVBRAIN enables incorporating accurate and probabilistic EEG forecasting into BCI applications, opening up new ways for investigating the functioning of the human brain.
Related
- Integration of vector and human behavior in residual malaria in rural communities in the Peruvian Amazon
- Retention-Constrained Post-Training Quantization of Cellpose-SAM for Stem Cell Microscopy
- DGM Is Dead; Long Live DGM-Star! A Quantization-Theoretic Approach to Gamma Rate Discretization
- Probabilistic inference of homonymous and heteronymous recurrent inhibition in human muscles from large-scale motor neuron recordings
- Probabilistic species distributions from nine large-scale gridded atlases over five decades
BibTeX
@misc{pankka2026revbrain,
title = {{REVBRAIN: Real-time Probabilistic EEG Forecasting Using Residual Vector Quantization}},
author = {Hanna Pankka and Jaakko Lehtinen and Timo Roine},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.38558},
eprint = {2610.38558},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.38558},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - REVBRAIN: Real-time Probabilistic EEG Forecasting Using Residual Vector Quantization
AU - Hanna Pankka
AU - Jaakko Lehtinen
AU - Timo Roine
PY - 2026
DA - 2026/10/28
PB - arXiv
DO - 10.48550/arxiv.2610.38558
UR - https://arxiv.org/abs/2610.38558
AB - Real-time forecasting of resting-state EEG signals is required by applications like brain-state-dependent stimulation, where computer generated actions are coupled with ongoing neural oscillations. Recently, the deep learning model WaveNet was proposed for the forecasting task. It shows superior accuracy compared to the prevailing method, the linear autoregressive (AR) model, and provides a built-in uncertainty measure due to its probabilistic modeling framework. However, inference using WaveNet is too slow for real-time use due to its sample-by-sample autoregressive operation: a single forecast requires evaluating the model tens or hundreds of times depending on the length of the prediction window. In this paper, we present REVBRAIN, a probabilistic deep learning method that produces forecasts in real time by generating the forecast over the whole prediction horizon at once. This is achieved by tokenizing the output space using residual vector quantization (RVQ) and generating forecasts over the items in the tokenized space. We tested REVBRAIN for forecasting 200 ms of EEG time series in the alpha-frequency band (8-13 Hz) and compared our results against WaveNet model. We found that REVBRAIN achieved real-time inference on a laptop GPU and an over 50× speedup compared to WaveNet, while reaching superior prediction accuracy. By achieving real-time inference speed, REVBRAIN enables incorporating accurate and probabilistic EEG forecasting into BCI applications, opening up new ways for investigating the functioning of the human brain.
ER -