Niklas Smedemark-Margulies
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
Niklas Smedemark-Margulies has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Deniz Erdogmus
Northeastern University, Boston, MA, USA
2 shared publications - Deniz Erdogmus
Northeastern University Boston MA USA
2 shared publications Deniz Erdoğmuş
Northeastern Univ, Dept Elect & Comp Engn, Boston, MA USA
2 shared publications- Eugene Tunik
Northeastern University
2 shared publications - Robin G. Walters
Clinical Trial Service Unit & Epidemiological Studies Unit (CTSU), Nuffield Department of Population Health, University of Oxford, Oxford, UK
2 shared publications - Ye Wang
Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA
2 shared publications
Latest Publications
- 1 February 2025
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 22 (1) , 016022
RSVP keyboard with inquiry preview: mixed performance and user experience with an adaptive, multimodal typing interface combining EEG and switch input
Objective. The RSVP Keyboard is a non-implantable, event-related potential-based brain-computer interface (BCI) system designed to support communication access for people with severe speech and physical impairments. Here we introduce inquiry preview (IP), a new RSVP Keyboard interface incorporating switch input for users with some voluntary motor function, and describe its effects on typing performance and other outcomes. Approach. Four individuals with disabilities participated in the collaborative design of possible switch input applications for the RSVP Keyboard, leading to the development of IP and a method of fusing switch input with language model and electroencephalography (EEG) evidence for typing. Twenty-four participants without disabilities and one potential end user with incomplete locked-in syndrome took part in two experiments investigating the effects of IP and two modes of switch input on typing accuracy and speed during a copy-spelling task. Main results. For participants without disabilities, IP and switch input tended to worsen typing performance compared to the standard RSVP Keyboard condition, with more consistent effects across participants for speed than for accuracy. However, there was considerable variability, with some participants demonstrating improved typing performance and better user experience (UX) with IP and switch input. Typing performance for the potential end user was comparable to that of participants without disabilities. He typed most quickly and accurately with IP and switch input and gave favorable UX ratings to those conditions, but preferred standard RSVP Keyboard. Significance. IP is a novel multimodal interface for the RSVP Keyboard BCI, incorporating switch input as an additional control signal. Typing performance and UX and preference varied widely across participants, reinforcing the need for flexible, customizable BCI systems that can adapt to individual users. ClinicalTrials.gov Identifier: NCT04468919.
- 1 February 2025
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 22 (1) , 016022
RSVP keyboard with inquiry preview: mixed performance and user experience with an adaptive, multimodal typing interface combining EEG and switch input
Objective. The RSVP Keyboard is a non-implantable, event-related potential-based brain-computer interface (BCI) system designed to support communication access for people with severe speech and physical impairments. Here we introduce inquiry preview (IP), a new RSVP Keyboard interface incorporating switch input for users with some voluntary motor function, and describe its effects on typing performance and other outcomes. Approach. Four individuals with disabilities participated in the collaborative design of possible switch input applications for the RSVP Keyboard, leading to the development of IP and a method of fusing switch input with language model and electroencephalography (EEG) evidence for typing. Twenty-four participants without disabilities and one potential end user with incomplete locked-in syndrome took part in two experiments investigating the effects of IP and two modes of switch input on typing accuracy and speed during a copy-spelling task. Main results. For participants without disabilities, IP and switch input tended to worsen typing performance compared to the standard RSVP Keyboard condition, with more consistent effects across participants for speed than for accuracy. However, there was considerable variability, with some participants demonstrating improved typing performance and better user experience (UX) with IP and switch input. Typing performance for the potential end user was comparable to that of participants without disabilities. He typed most quickly and accurately with IP and switch input and gave favorable UX ratings to those conditions, but preferred standard RSVP Keyboard. Significance. IP is a novel multimodal interface for the RSVP Keyboard BCI, incorporating switch input as an additional control signal. Typing performance and UX and preference varied widely across participants, reinforcing the need for flexible, customizable BCI systems that can adapt to individual users. ClinicalTrials.gov Identifier: NCT04468919.
- 1 December 2024
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 21 (6) , 066031
Improving subject transfer in EEG classification with divergence estimation
Objective. Classification models for electroencephalogram (EEG) data show a large decrease in performance when evaluated on unseen test subjects. We improve performance using new regularization techniques during model training. Approach. We propose several graphical models to describe an EEG classification task. From each model, we identify statistical relationships that should hold true in an idealized training scenario (with infinite data and a globally-optimal model) but that may not hold in practice. We design regularization penalties to enforce these relationships in two stages. First, we identify suitable proxy quantities (divergences such as Mutual Information and Wasserstein-1) that can be used to measure statistical independence and dependence relationships. Second, we provide algorithms to efficiently estimate these quantities during training using secondary neural network models. Main results. We conduct extensive computational experiments using a large benchmark EEG dataset, comparing our proposed techniques with a baseline method that uses an adversarial classifier. We first show the performance of each method across a wide range of hyperparameters, demonstrating that each method can be easily tuned to yield significant benefits over an unregularized model. We show that, using ideal hyperparameters for all methods, our first technique gives significantly better performance than the baseline regularization technique. We also show that, across hyperparameters, our second technique gives significantly more stable performance than the baseline. The proposed methods require only a small computational cost at training time that is equivalent to the cost of the baseline. Significance. The high variability in signal distribution between subjects means that typical approaches to EEG signal modeling often require time-intensive calibration for each user, and even re-calibration before every use. By improving the performance of population models in the most stringent case of zero-shot subject transfer, we may help reduce or eliminate the need for model calibration.
- 1 December 2024
- Published by IOP Publishing in Journal of Neural Engineering
- Vol. 21 (6) , 066031
Improving subject transfer in EEG classification with divergence estimation
Objective. Classification models for electroencephalogram (EEG) data show a large decrease in performance when evaluated on unseen test subjects. We improve performance using new regularization techniques during model training. Approach. We propose several graphical models to describe an EEG classification task. From each model, we identify statistical relationships that should hold true in an idealized training scenario (with infinite data and a globally-optimal model) but that may not hold in practice. We design regularization penalties to enforce these relationships in two stages. First, we identify suitable proxy quantities (divergences such as Mutual Information and Wasserstein-1) that can be used to measure statistical independence and dependence relationships. Second, we provide algorithms to efficiently estimate these quantities during training using secondary neural network models. Main results. We conduct extensive computational experiments using a large benchmark EEG dataset, comparing our proposed techniques with a baseline method that uses an adversarial classifier. We first show the performance of each method across a wide range of hyperparameters, demonstrating that each method can be easily tuned to yield significant benefits over an unregularized model. We show that, using ideal hyperparameters for all methods, our first technique gives significantly better performance than the baseline regularization technique. We also show that, across hyperparameters, our second technique gives significantly more stable performance than the baseline. The proposed methods require only a small computational cost at training time that is equivalent to the cost of the baseline. Significance. The high variability in signal distribution between subjects means that typical approaches to EEG signal modeling often require time-intensive calibration for each user, and even re-calibration before every use. By improving the performance of population models in the most stringent case of zero-shot subject transfer, we may help reduce or eliminate the need for model calibration.
- Open Access
- 1 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Systems and Rehabilitation Engineering
- Vol. 32 (15344320) , 1187-1197
User Training With Error Augmentation for sEMG-Based Gesture Classification
We designed and tested a system for real-time control of a user interface by extracting surface electromyographic (sEMG) activity from eight electrodes in a wristband configuration. sEMG data were streamed into a machine-learning algorithm that classified hand gestures in real-time. After an initial model calibration, participants were presented with one of three types of feedback during a human-learning stage: veridical feedback, in which predicted probabilities from the gesture classification algorithm were displayed without alteration; modified feedback, in which we applied a hidden augmentation of error to these probabilities; and no feedback. User performance was then evaluated in a series of minigames, in which subjects were required to use eight gestures to manipulate their game avatar to complete a task. Experimental results indicated that relative to the baseline, the modified feedback condition led to significantly improved accuracy. Class separation also improved, though this trend was not significant. These findings suggest that real-time feedback in a gamified user interface with manipulation of feedback may enable intuitive, rapid, and accurate task acquisition for sEMG-based gesture recognition applications.
- 1 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Systems and Rehabilitation Engineering
- Vol. 32 (15344320) , 1187-1197
Open AccessUser Training With Error Augmentation for sEMG-Based Gesture Classification
We designed and tested a system for real-time control of a user interface by extracting surface electromyographic (sEMG) activity from eight electrodes in a wristband configuration. sEMG data were streamed into a machine-learning algorithm that classified hand gestures in real-time. After an initial model calibration, participants were presented with one of three types of feedback during a human-learning stage: veridical feedback, in which predicted probabilities from the gesture classification algorithm were displayed without alteration; modified feedback, in which we applied a hidden augmentation of error to these probabilities; and no feedback. User performance was then evaluated in a series of minigames, in which subjects were required to use eight gestures to manipulate their game avatar to complete a task. Experimental results indicated that relative to the baseline, the modified feedback condition led to significantly improved accuracy. Class separation also improved, though this trend was not significant. These findings suggest that real-time feedback in a gamified user interface with manipulation of feedback may enable intuitive, rapid, and accurate task acquisition for sEMG-based gesture recognition applications.
- Open Access
Fast and Expressive Gesture Recognition using a Combination-Homomorphic Electromyogram Encoder
We study the task of gesture recognition from electromyography (EMG), with the goal of enabling expressive human-computer interaction at high accuracy, while minimizing the time required for new subjects to provide calibration data. To fulfill these goals, we define combination gestures consisting of a direction component and a modifier component. New subjects only demonstrate the single component gestures and we seek to extrapolate from these to all possible single or combination gestures. We extrapolate to unseen combination gestures by combining the feature vectors of real single gestures to produce synthetic training data. This strategy allows us to provide a large and flexible gesture vocabulary, while not requiring new subjects to demonstrate combinatorially many example gestures. We pre-train an encoder and a combination operator using self-supervision, so that we can produce useful synthetic training data for unseen test subjects. To evaluate the proposed method, we collect a real-world EMG dataset, and measure the effect of augmented supervision against two baselines: a partially-supervised model trained with only single gesture data from the unseen subject, and a fully-supervised model trained with real single and real combination gesture data from the unseen subject. We find that the proposed method provides a dramatic improvement over the partially-supervised model, and achieves a useful classification accuracy that in some cases approaches the performance of the fully-supervised model.
Open AccessFast and Expressive Gesture Recognition using a Combination-Homomorphic Electromyogram Encoder
We study the task of gesture recognition from electromyography (EMG), with the goal of enabling expressive human-computer interaction at high accuracy, while minimizing the time required for new subjects to provide calibration data. To fulfill these goals, we define combination gestures consisting of a direction component and a modifier component. New subjects only demonstrate the single component gestures and we seek to extrapolate from these to all possible single or combination gestures. We extrapolate to unseen combination gestures by combining the feature vectors of real single gestures to produce synthetic training data. This strategy allows us to provide a large and flexible gesture vocabulary, while not requiring new subjects to demonstrate combinatorially many example gestures. We pre-train an encoder and a combination operator using self-supervision, so that we can produce useful synthetic training data for unseen test subjects. To evaluate the proposed method, we collect a real-world EMG dataset, and measure the effect of augmented supervision against two baselines: a partially-supervised model trained with only single gesture data from the unseen subject, and a fully-supervised model trained with real single and real combination gesture data from the unseen subject. We find that the proposed method provides a dramatic improvement over the partially-supervised model, and achieves a useful classification accuracy that in some cases approaches the performance of the fully-supervised model.
- Open Access
- 7 November 2022
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 18 (11) , e1010591
Probabilistic program inference in network-based epidemiological simulations
Accurate epidemiological models require parameter estimates that account for mobility patterns and social network structure. We demonstrate the effectiveness of probabilistic programming for parameter inference in these models. We consider an agent-based simulation that represents mobility networks as degree-corrected stochastic block models, whose parameters we estimate from cell phone co-location data. We then use probabilistic program inference methods to approximate the distribution over disease transmission parameters conditioned on reported cases and deaths. Our experiments demonstrate that the resulting models improve the quality of fit in multiple geographies relative to baselines that do not model network topology.
- 7 November 2022
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 18 (11) , e1010591
Open AccessProbabilistic program inference in network-based epidemiological simulations
Accurate epidemiological models require parameter estimates that account for mobility patterns and social network structure. We demonstrate the effectiveness of probabilistic programming for parameter inference in these models. We consider an agent-based simulation that represents mobility networks as degree-corrected stochastic block models, whose parameters we estimate from cell phone co-location data. We then use probabilistic program inference methods to approximate the distribution over disease transmission parameters conditioned on reported cases and deaths. Our experiments demonstrate that the resulting models improve the quality of fit in multiple geographies relative to baselines that do not model network topology.
- Open Access
AutoTransfer: Subject Transfer Learning with Censored Representations on Biosignals Data
We provide a regularization framework for subject transfer learning in which we seek to train an encoder and classifier to minimize classification loss, subject to a penalty measuring independence between the latent representation and the subject label. We introduce three notions of independence and corresponding penalty terms using mutual information or divergence as a proxy for independence. For each penalty term, we provide several concrete estimation algorithms, using analytic methods as well as neural critic functions. We provide a hands-off strategy for applying this diverse family of regularization algorithms to a new dataset, which we call "AutoTransfer". We evaluate the performance of these individual regularization strategies and our AutoTransfer method on EEG, EMG, and ECoG datasets, showing that these approaches can improve subject transfer learning for challenging real-world datasets.
Open AccessAutoTransfer: Subject Transfer Learning with Censored Representations on Biosignals Data
We provide a regularization framework for subject transfer learning in which we seek to train an encoder and classifier to minimize classification loss, subject to a penalty measuring independence between the latent representation and the subject label. We introduce three notions of independence and corresponding penalty terms using mutual information or divergence as a proxy for independence. For each penalty term, we provide several concrete estimation algorithms, using analytic methods as well as neural critic functions. We provide a hands-off strategy for applying this diverse family of regularization algorithms to a new dataset, which we call "AutoTransfer". We evaluate the performance of these individual regularization strategies and our AutoTransfer method on EEG, EMG, and ECoG datasets, showing that these approaches can improve subject transfer learning for challenging real-world datasets.
- 9 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 28 (10709908) , 867-871
Geometric Analysis of Uncertainty Sampling for Dense Neural Network Layer
For model adaptation of fully connected neural network layers, we provide an information geometric and sample behavioral active learning uncertainty sampling objective analysis. We identify conditions under which several uncertainty-based methods have the same performance and show that such conditions are more likely to appear in the early stages of learning. We define riskier samples for adaptation, and demonstrate that, as the set of labeled samples increases, margin-based sampling outperforms other uncertainty sampling methods by preferentially selecting these risky samples. We support our derivations and illustrations with experiments using Meta-Dataset, a benchmark for few-shot learning. We compare uncertainty-based active learning objectives using features produced by SimpleCNAPS (a state-of-the-art few-shot classifier) as input for a fully-connected adaptation layer. Our results indicate that margin-based uncertainty sampling achieves similar performance as other uncertainty based sampling methods with fewer labelled samples as discussed in the novel geometric analysis.
- 9 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 28 (10709908) , 867-871
Geometric Analysis of Uncertainty Sampling for Dense Neural Network Layer
For model adaptation of fully connected neural network layers, we provide an information geometric and sample behavioral active learning uncertainty sampling objective analysis. We identify conditions under which several uncertainty-based methods have the same performance and show that such conditions are more likely to appear in the early stages of learning. We define riskier samples for adaptation, and demonstrate that, as the set of labeled samples increases, margin-based sampling outperforms other uncertainty sampling methods by preferentially selecting these risky samples. We support our derivations and illustrations with experiments using Meta-Dataset, a benchmark for few-shot learning. We compare uncertainty-based active learning objectives using features produced by SimpleCNAPS (a state-of-the-art few-shot classifier) as input for a fully-connected adaptation layer. Our results indicate that margin-based uncertainty sampling achieves similar performance as other uncertainty based sampling methods with fewer labelled samples as discussed in the novel geometric analysis.