Cortically Coupled Computing: A New Paradigm for Synergistic Human-Machine Interaction

Saproo, Sameer; Faller, Josef; Shih, Victor; Sajda, Paul; Waytowich, Nicholas R.; Bohannon, Addison; Lawhern, Vernon J.; Lance, Brent J.; Jangraw, David C.

doi:10.1109/mc.2016.294

Cited by 14 publications

(20 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…are much larger (see Section 2.3 for more details on our within-and cross-subject analyses). We apply an average pooling layer of size (1,4) to reduce the sampling rate of the signal to 32Hz. We also regularize each spatial filter by using a maximum norm constraint of 1 on its weights; w 2 < 1.…”

Section: Eegnet: Compact Cnn Architecturementioning

confidence: 99%

“…(1) 0.8866 (2) 0.9076 (3) 0.8910 (4) 0.8747 (1,2) 0.8875 (1,3) 0.8593 (1,4) 0.8325 (2,3) 0.8923 (2,4) 0.8721 (3,4) 0.8206 (1,2,3) 0.8637 (1,2,4) 0.8202 (1,3,4) 0.7108 (2,3,4) 0.7970 None 0.9054 Table 4: Performance of a cross-subject trained EEGNet-4,1 model when removing certain filters from the model, then using the model to predict the test set for one randomly chosen fold of the P300 dataset. AUC values in bold denote the best performing model when removing 1, 2 or 3 filters at a time.…”

Section: Filters Removed Test Set Aucmentioning

confidence: 99%

See 1 more Smart Citation

EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces

et al. 2018

View full text Add to dashboard Cite

show abstract

Section: Eegnet: Compact Cnn Architecturementioning

confidence: 99%

Section: Filters Removed Test Set Aucmentioning

confidence: 99%

EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Finally, as our decoding approach (1) can be pre-trained, (2) is capable of ignoring co-morbid, experimentally induced components or artifacts, and (3) exhibits a certain amount of invariance to temporal uncertainty, this work represents an important extension beyond the traditional approaches employed in the measurement and interpretation of neural phenomena. We believe the use of such CNN-based decoding will enable more complex, real-world neuroscience research [55][56][57]. By allowing the experimenter to develop decoding models from precisely controlled laboratory experiments, yet analyze data from a much smaller number of trials without requiring the same level of temporal precision, CNNs such as the one presented here, can help bridge the gap between our knowledge of how the brain functions in the laboratory and how it may function in the real-world.…”

Section: Discussionmentioning

confidence: 99%

Decoding P300 Variability using Convolutional Neural Networks

Solon

Lawhern

Touryan

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Deep convolutional neural networks (CNN) have previously been shown to be useful tools for signal decoding and analysis in a variety of complex domains, such as image processing and speech recognition. By learning from large amounts of data, the representations encoded by these deep networks are often invariant to moderate changes in the underlying feature spaces. Recently, we proposed a CNN architecture that could be applied to electroencephalogram (EEG) decoding and analysis. In this article, we train our CNN model using data from prior experiments in order to later decode the P300 evoked response from an unseen, hold-out experiment. We analyze the CNN output as a function of the underlying variability in the P300 response and demonstrate that the CNN output is sensitive to the experiment-induced changes in the neural response. We then assess the utility of our approach as a means of improving the overall signalto-noise ratio in the EEG record. Finally, we show an example of how CNN-based decoding can be applied to the analysis of complex data.

show abstract

“…Human augmentation aims at extending human cognitive abilities, often in a situated, task-specific fashion. Previous research has demonstrated through various implemented prototypes and experiments the feasibility of extending human perceptive abilities or information processing and decision-making abilities [ 1 , 2 ]. In the latter case, Artificial Intelligence (AI) techniques are poised to play a significant role in providing the task-specific information processing power supporting the augmentation aspects.…”

Section: Introduction and Rationalementioning

confidence: 99%

“…Although human augmentation systems have been developed prior to the popularisation of Brain–Computer Interfaces (BCI), these have taken a more prominent role in recent years, as they offer a seamless mechanism to capture elements of human cognitive processes in a way that enables the synchronisation of computations [ 1 , 2 ]. With the rise of autonomous intelligent systems, a new application of human augmentation has been suggested in order to keep humans in control of autonomous AI systems whose performance could potentially exceed even that of human experts.…”

Section: Introduction and Rationalementioning

confidence: 99%

A Motivational Model of BCI-Controlled Heuristic Search

Cavazza¹

2018

Brain Sciences

View full text Add to dashboard Cite

Several researchers have proposed a new application for human augmentation, which is to provide human supervision to autonomous artificial intelligence (AI) systems. In this paper, we introduce a framework to implement this proposal, which consists of using Brain–Computer Interfaces (BCI) to influence AI computation via some of their core algorithmic components, such as heuristic search. Our framework is based on a joint analysis of philosophical proposals characterising the behaviour of autonomous AI systems and recent research in cognitive neuroscience that support the design of appropriate BCI. Our framework is defined as a motivational approach, which, on the AI side, influences the shape of the solution produced by heuristic search using a BCI motivational signal reflecting the user’s disposition towards the anticipated result. The actual mapping is based on a measure of prefrontal asymmetry, which is translated into a non-admissible variant of the heuristic function. Finally, we discuss results from a proof-of-concept experiment using functional near-infrared spectroscopy (fNIRS) to capture prefrontal asymmetry and control the progression of AI computation of traditional heuristic search problems.

show abstract

Cortically Coupled Computing: A New Paradigm for Synergistic Human-Machine Interaction

Cited by 14 publications

References 12 publications

EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces

EEGNet: a compact convolutional neural network for EEG-based brain–computer interfaces

Decoding P300 Variability using Convolutional Neural Networks

A Motivational Model of BCI-Controlled Heuristic Search

Contact Info

Product

Resources

About