Prof. Dr. rer. nat. Ioannis Iossifidis

AUSGEWÄHLTE PUBLIKATIONEN

• anthropomorphic robot arm anthropomorphism arm movement model artificial skin attractor dynamics approach Autonomous robotics BCI behavior generation direct physical interaction Dynamical systems haptic interface human anatomy human motor behavior human operators intuition inverse kinematics Machine Learning man machine interaction man-machine systems man-machine-interaction manipulators mechanical structure modeling movement model object recognition operators gesture pervasive design robot manipulator control robot vision robotic assistant sensory channels user modelling

  Zeige alle

  79 Einträge « ‹ 1 von 8 › »

  2023

  79.

  Ali, Omair; Saif-ur-Rehman, Muhammad; Glasmachers, Tobias; Iossifidis, Ioannis; Klaes, Christian

  ConTraNet: A Hybrid Network for Improving the Classification of EEG and EMG Signals with Limited Training Data Artikel

  In: Computers in Biology and Medicine, S. 107649, 2023, ISSN: 0010-4825.

  Abstract | Links | BibTeX | Schlagwörter: BCI, Brain computer interface, Deep learning, EEG decoding, EMG decoding, Machine Learning

  @article{aliConTraNetHybridNetwork2023,
  
  title = {ConTraNet: A Hybrid Network for Improving the Classification of EEG and EMG Signals with Limited Training Data},
  
  author = {Omair Ali and Muhammad Saif-ur-Rehman and Tobias Glasmachers and Ioannis Iossifidis and Christian Klaes},
  
  url = {https://www.sciencedirect.com/science/article/pii/S0010482523011149},
  
  doi = {10.1016/j.compbiomed.2023.107649},
  
  issn = {0010-4825},
  
  year  = {2023},
  
  date = {2023-11-02},
  
  urldate = {2023-11-02},
  
  journal = {Computers in Biology and Medicine},
  
  pages = {107649},
  
  abstract = {Objective Bio-Signals such as electroencephalography (EEG) and electromyography (EMG) are widely used for the rehabilitation of physically disabled people and for the characterization of cognitive impairments. Successful decoding of these bio-signals is however non-trivial because of the time-varying and non-stationary characteristics. Furthermore, existence of short- and long-range dependencies in these time-series signal makes the decoding even more challenging. State-of-the-art studies proposed Convolutional Neural Networks (CNNs) based architectures for the classification of these bio-signals, which are proven useful to learn spatial representations. However, CNNs because of the fixed size convolutional kernels and shared weights pay only uniform attention and are also suboptimal in learning short-long term dependencies, simultaneously, which could be pivotal in decoding EEG and EMG signals. Therefore, it is important to address these limitations of CNNs. To learn short- and long-range dependencies simultaneously and to pay more attention to more relevant part of the input signal, Transformer neural network-based architectures can play a significant role. Nonetheless, it requires a large corpus of training data. However, EEG and EMG decoding studies produce limited amount of the data. Therefore, using standalone transformers neural networks produce ordinary results. In this study, we ask a question whether we can fix the limitations of CNN and transformer neural networks and provide a robust and generalized model that can simultaneously learn spatial patterns, long-short term dependencies, pay variable amount of attention to time-varying non-stationary input signal with limited training data. Approach In this work, we introduce a novel single hybrid model called ConTraNet, which is based on CNN and Transformer architectures that contains the strengths of both CNN and Transformer neural networks. ConTraNet uses a CNN block to introduce inductive bias in the model and learn local dependencies, whereas the Transformer block uses the self-attention mechanism to learn the short- and long-range or global dependencies in the signal and learn to pay different attention to different parts of the signals. Main results We evaluated and compared the ConTraNet with state-of-the-art methods on four publicly available datasets (BCI Competition IV dataset 2b, Physionet MI-EEG dataset, Mendeley sEMG dataset, Mendeley sEMG V1 dataset) which belong to EEG-HMI and EMG-HMI paradigms. ConTraNet outperformed its counterparts in all the different category tasks (2-class, 3-class, 4-class, 7-class, and 10-class decoding tasks). Significance With limited training data ConTraNet significantly improves classification performance on four publicly available datasets for 2, 3, 4, 7, and 10-classes compared to its counterparts.},
  
  keywords = {BCI, Brain computer interface, Deep learning, EEG decoding, EMG decoding, Machine Learning},
  
  pubstate = {published},
  
  tppubtype = {article}
  
  }
  
  

  Schließen

  Objective Bio-Signals such as electroencephalography (EEG) and electromyography (EMG) are widely used for the rehabilitation of physically disabled people and for the characterization of cognitive impairments. Successful decoding of these bio-signals is however non-trivial because of the time-varying and non-stationary characteristics. Furthermore, existence of short- and long-range dependencies in these time-series signal makes the decoding even more challenging. State-of-the-art studies proposed Convolutional Neural Networks (CNNs) based architectures for the classification of these bio-signals, which are proven useful to learn spatial representations. However, CNNs because of the fixed size convolutional kernels and shared weights pay only uniform attention and are also suboptimal in learning short-long term dependencies, simultaneously, which could be pivotal in decoding EEG and EMG signals. Therefore, it is important to address these limitations of CNNs. To learn short- and long-range dependencies simultaneously and to pay more attention to more relevant part of the input signal, Transformer neural network-based architectures can play a significant role. Nonetheless, it requires a large corpus of training data. However, EEG and EMG decoding studies produce limited amount of the data. Therefore, using standalone transformers neural networks produce ordinary results. In this study, we ask a question whether we can fix the limitations of CNN and transformer neural networks and provide a robust and generalized model that can simultaneously learn spatial patterns, long-short term dependencies, pay variable amount of attention to time-varying non-stationary input signal with limited training data. Approach In this work, we introduce a novel single hybrid model called ConTraNet, which is based on CNN and Transformer architectures that contains the strengths of both CNN and Transformer neural networks. ConTraNet uses a CNN block to introduce inductive bias in the model and learn local dependencies, whereas the Transformer block uses the self-attention mechanism to learn the short- and long-range or global dependencies in the signal and learn to pay different attention to different parts of the signals. Main results We evaluated and compared the ConTraNet with state-of-the-art methods on four publicly available datasets (BCI Competition IV dataset 2b, Physionet MI-EEG dataset, Mendeley sEMG dataset, Mendeley sEMG V1 dataset) which belong to EEG-HMI and EMG-HMI paradigms. ConTraNet outperformed its counterparts in all the different category tasks (2-class, 3-class, 4-class, 7-class, and 10-class decoding tasks). Significance With limited training data ConTraNet significantly improves classification performance on four publicly available datasets for 2, 3, 4, 7, and 10-classes compared to its counterparts.

  Schließen

  • https://www.sciencedirect.com/science/article/pii/S0010482523011149
  • doi:10.1016/j.compbiomed.2023.107649

  Schließen

  78.

  Fidencio, Aline Xavier; Klaes, Christian; Iossifidis, Ioannis

  Exploring Error-related Potentials in Adaptive Brain-Machine Interfaces: Challenges and Investigation of Occurrence and Detection Ratios Proceedings Article

  In: BC23 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022, BCCN Bernstein Network Computational Network, 2023.

  Abstract | BibTeX | Schlagwörter: BCI, EEG, Machine Learning

  @inproceedings{xavierfidencioExploringErrorrelatedPotentials2023,
  
  title = {Exploring Error-related Potentials in Adaptive Brain-Machine Interfaces: Challenges and Investigation of Occurrence and Detection Ratios},
  
  author = {Aline Xavier Fidencio and Christian Klaes and Ioannis Iossifidis},
  
  year  = {2023},
  
  date = {2023-09-15},
  
  urldate = {2023-09-15},
  
  booktitle = {BC23 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022},
  
  publisher = {BCCN Bernstein Network Computational Network},
  
  abstract = {Non-invasive techniques like EEG can record error-related potentials (ErrPs), neural signals associated with error processing and awareness. ErrPs are generated in response to self-made and external errors, including those produced by the BMI. Since ErrPs are implicitly elicited and don’t add extra workload for the subject, they serve as a natural and intrinsic feedback source for developing adaptive BMIs. In our study, we assess the occurrence of interaction ErrPs in an adaptive BMI that combines ErrPs and reinforcement learning. We intentionally provoke ErrPs when the BMI misinterprets the user’s intention and performs an incorrect action. Subjects participated in a game controlled by a keyboard and/or motor imagery (imagining hand movements), and EEG data were recorded using an eight-electrode gel-based EEG system. Results reveal that obtaining a distinct ErrPs signal for each subject is more challenging than anticipated. Current practices report the ErrP in terms of over all subjects and trials difference grand average (error minus correct). This approach has, however, the limitation of masking the inter-trial and subject variability, which are relevant for the online single-trial detection of such signals. Moreover, the reported ErrPs waveshape exhibit differences in terms of components observed, as well as their respective latencies, even when very similar tasks are used. Consequently, we conducted additional individualized data analysis to gain deeper insights into the single-trial nature of the ErrPs. As a result, we determined the need for a better understanding and further investigation of how effectively the ErrPs waveforms generalize across subjects, tasks, experimental protocols, and feedback modalities. Given the challenges in obtaining a clear signal for all subjects and the limitations found in existing literature (Xavier Fidêncio et al., 2022), we hypothesize whether an error signal measurable at the scalp level is consistently generated when subjects encounter erroneous conditions. To address this question, we will assess the occurrence-to-detection ratio of ErrPs using invasive and non-invasive recording techniques, examining how uncertainties regarding error generation in the brain impact the learning pipeline.},
  
  keywords = {BCI, EEG, Machine Learning},
  
  pubstate = {published},
  
  tppubtype = {inproceedings}
  
  }
  
  

  Schließen

  Non-invasive techniques like EEG can record error-related potentials (ErrPs), neural signals associated with error processing and awareness. ErrPs are generated in response to self-made and external errors, including those produced by the BMI. Since ErrPs are implicitly elicited and don’t add extra workload for the subject, they serve as a natural and intrinsic feedback source for developing adaptive BMIs. In our study, we assess the occurrence of interaction ErrPs in an adaptive BMI that combines ErrPs and reinforcement learning. We intentionally provoke ErrPs when the BMI misinterprets the user’s intention and performs an incorrect action. Subjects participated in a game controlled by a keyboard and/or motor imagery (imagining hand movements), and EEG data were recorded using an eight-electrode gel-based EEG system. Results reveal that obtaining a distinct ErrPs signal for each subject is more challenging than anticipated. Current practices report the ErrP in terms of over all subjects and trials difference grand average (error minus correct). This approach has, however, the limitation of masking the inter-trial and subject variability, which are relevant for the online single-trial detection of such signals. Moreover, the reported ErrPs waveshape exhibit differences in terms of components observed, as well as their respective latencies, even when very similar tasks are used. Consequently, we conducted additional individualized data analysis to gain deeper insights into the single-trial nature of the ErrPs. As a result, we determined the need for a better understanding and further investigation of how effectively the ErrPs waveforms generalize across subjects, tasks, experimental protocols, and feedback modalities. Given the challenges in obtaining a clear signal for all subjects and the limitations found in existing literature (Xavier Fidêncio et al., 2022), we hypothesize whether an error signal measurable at the scalp level is consistently generated when subjects encounter erroneous conditions. To address this question, we will assess the occurrence-to-detection ratio of ErrPs using invasive and non-invasive recording techniques, examining how uncertainties regarding error generation in the brain impact the learning pipeline.

  Schließen

  77.

  Grün, Felix; Iossifidis, Ioannis

  Investigation of the Interplay of Model-Based and Model-Free Learning Using Reinforcement Learning Proceedings Article

  In: BC23 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022, BCCN Bernstein Network Computational Network, 2023.

  Abstract | BibTeX | Schlagwörter: Machine Learning, Reinforcement learning

  @inproceedings{grunInvestigationInterplayModelBased2023,
  
  title = {Investigation of the Interplay of Model-Based and Model-Free Learning Using Reinforcement Learning},
  
  author = {Felix Grün and Ioannis Iossifidis},
  
  year  = {2023},
  
  date = {2023-09-15},
  
  urldate = {2023-09-15},
  
  booktitle = {BC23 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022},
  
  publisher = {BCCN Bernstein Network Computational Network},
  
  abstract = {The reward prediction error hypothesis of dopamine in the brain states that activity of dopaminergic neurons in certain brain regions correlates with the reward prediction error that corresponds to the temporal difference error, often used as a learning signal in model free reinforcement learning (RL). This suggests that some form of reinforcement learning is used in animal and human brains when learning a task. On the other hand, it is clear that humans are capable of building an internal model of a task, or environment, and using it for planning, especially in sequential tasks. In RL, these two learning approaches, model-driven and reward-driven, are known as model based and model-free RL approaches. Both systems were previously thought to exist in parallel, with some higher process choosing which to use. A decade ago, research suggested both could be used concurrently, with some subject-specific weight assigned to each [1]. Still, the prevalent belief appeared to be that model-free learning is the default mechanism used, replaced or assisted by model-based planning only when the task demands it, i.e. higher rewards justify the additional cognitive effort. Recently, Feher da Silva et al. [2] questioned this belief, presenting data and analyses that indicate model-based learning may be used on its own and can even be computationally more efficient. We take a RL perspective, consider different ways to combine model-based and model-free approaches for modeling and for performance, and discuss how to further study this interplay in human behavioral experiments.},
  
  keywords = {Machine Learning, Reinforcement learning},
  
  pubstate = {published},
  
  tppubtype = {inproceedings}
  
  }
  
  

  Schließen

  The reward prediction error hypothesis of dopamine in the brain states that activity of dopaminergic neurons in certain brain regions correlates with the reward prediction error that corresponds to the temporal difference error, often used as a learning signal in model free reinforcement learning (RL). This suggests that some form of reinforcement learning is used in animal and human brains when learning a task. On the other hand, it is clear that humans are capable of building an internal model of a task, or environment, and using it for planning, especially in sequential tasks. In RL, these two learning approaches, model-driven and reward-driven, are known as model based and model-free RL approaches. Both systems were previously thought to exist in parallel, with some higher process choosing which to use. A decade ago, research suggested both could be used concurrently, with some subject-specific weight assigned to each [1]. Still, the prevalent belief appeared to be that model-free learning is the default mechanism used, replaced or assisted by model-based planning only when the task demands it, i.e. higher rewards justify the additional cognitive effort. Recently, Feher da Silva et al. [2] questioned this belief, presenting data and analyses that indicate model-based learning may be used on its own and can even be computationally more efficient. We take a RL perspective, consider different ways to combine model-based and model-free approaches for modeling and for performance, and discuss how to further study this interplay in human behavioral experiments.

  Schließen

  76.

  Schmidt, Marie Dominique; Iossifidis, Ioannis

  The Link between Muscle Activity and Upper Limb Kinematics Proceedings Article

  In: BC23 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022, BCCN Bernstein Network Computational Network, 2023.

  Abstract | BibTeX | Schlagwörter: BCI, Machine Learning

  @inproceedings{schmidtLinkMuscleActivity2023,
  
  title = {The Link between Muscle Activity and Upper Limb Kinematics},
  
  author = {Marie Dominique Schmidt and Ioannis Iossifidis},
  
  year  = {2023},
  
  date = {2023-09-15},
  
  urldate = {2023-09-15},
  
  booktitle = {BC23 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022},
  
  publisher = {BCCN Bernstein Network Computational Network},
  
  abstract = {The upper limbs are crucial in performing daily tasks that require strength, a wide range of motion, and precision. To achieve coordinated motion, planning and timing are critical. Sensory information about the target and the current body state is essential, as well as integrating past experiences, represented by pre-learned inverse dynamics that generate associated muscle activity. We propose a generative model that predicts upper limb muscle activity from a variety of simple and complex everyday motions by means of a recurrent neural network. The model shows promising results, with a good fit for different subjects and abstracts well for new motions. We handle the high inter-subject variation in muscle activity using a transfer learning approach, resulting in a good fit for new subjects. Our approach has implications for fundamental movement control understanding and the rehabilitation of neuromuscular diseases using myoelectric prostheses and functional electrical stimulation. Our model can efficiently predict both muscle activity and motion trajectory, which can assist in developing more effective rehabilitation techniques.},
  
  keywords = {BCI, Machine Learning},
  
  pubstate = {published},
  
  tppubtype = {inproceedings}
  
  }
  
  

  Schließen

  The upper limbs are crucial in performing daily tasks that require strength, a wide range of motion, and precision. To achieve coordinated motion, planning and timing are critical. Sensory information about the target and the current body state is essential, as well as integrating past experiences, represented by pre-learned inverse dynamics that generate associated muscle activity. We propose a generative model that predicts upper limb muscle activity from a variety of simple and complex everyday motions by means of a recurrent neural network. The model shows promising results, with a good fit for different subjects and abstracts well for new motions. We handle the high inter-subject variation in muscle activity using a transfer learning approach, resulting in a good fit for new subjects. Our approach has implications for fundamental movement control understanding and the rehabilitation of neuromuscular diseases using myoelectric prostheses and functional electrical stimulation. Our model can efficiently predict both muscle activity and motion trajectory, which can assist in developing more effective rehabilitation techniques.

  Schließen

  75.

  Schmidt, Marie D.; Glasmachers, Tobias; Iossifidis, Ioannis

  The Concepts of Muscle Activity Generation Driven by Upper Limb Kinematics Artikel

  In: BioMedical Engineering OnLine, Bd. 22, Nr. 1, S. 63, 2023, ISSN: 1475-925X.

  Abstract | Links | BibTeX | Schlagwörter: Artificial generated signal, BCI, Electromyography (EMG), Generative model, Inertial measurement unit (IMU), Machine Learning, Motion parameters, Muscle activity, Neural networks, transfer learning, Voluntary movement

  @article{schmidtConceptsMuscleActivity2023,
  
  title = {The Concepts of Muscle Activity Generation Driven by Upper Limb Kinematics},
  
  author = {Marie D. Schmidt and Tobias Glasmachers and Ioannis Iossifidis},
  
  url = {https://doi.org/10.1186/s12938-023-01116-9},
  
  doi = {10.1186/s12938-023-01116-9},
  
  issn = {1475-925X},
  
  year  = {2023},
  
  date = {2023-06-24},
  
  urldate = {2023-06-24},
  
  journal = {BioMedical Engineering OnLine},
  
  volume = {22},
  
  number = {1},
  
  pages = {63},
  
  abstract = {The underlying motivation of this work is to demonstrate that artificial muscle activity of known and unknown motion can be generated based on motion parameters, such as angular position, acceleration, and velocity of each joint (or the end-effector instead), which are similarly represented in our brains. This model is motivated by the known motion planning process in the central nervous system. That process incorporates the current body state from sensory systems and previous experiences, which might be represented as pre-learned inverse dynamics that generate associated muscle activity.},
  
  keywords = {Artificial generated signal, BCI, Electromyography (EMG), Generative model, Inertial measurement unit (IMU), Machine Learning, Motion parameters, Muscle activity, Neural networks, transfer learning, Voluntary movement},
  
  pubstate = {published},
  
  tppubtype = {article}
  
  }
  
  

  Schließen

  The underlying motivation of this work is to demonstrate that artificial muscle activity of known and unknown motion can be generated based on motion parameters, such as angular position, acceleration, and velocity of each joint (or the end-effector instead), which are similarly represented in our brains. This model is motivated by the known motion planning process in the central nervous system. That process incorporates the current body state from sensory systems and previous experiences, which might be represented as pre-learned inverse dynamics that generate associated muscle activity.

  Schließen

  • https://doi.org/10.1186/s12938-023-01116-9
  • doi:10.1186/s12938-023-01116-9

  Schließen

  74.

  Saif-ur-Rehman, Muhammad; Ali, Omair; Klaes, Christian; Iossifidis, Ioannis

  Adaptive SpikeDeep-Classifier: Self-organizing and self-supervised machine learning algorithm for online spike sorting Artikel

  In: arXiv:2304.01355 [cs, math, q-bio], 2023.

  Links | BibTeX | Schlagwörter: BCI, Machine Learning, Spike Sorting

  @article{saifurrehman2023adaptive,
  
  title = {Adaptive SpikeDeep-Classifier: Self-organizing and self-supervised machine learning algorithm for online spike sorting},
  
  author = {Muhammad Saif-ur-Rehman and Omair Ali and Christian Klaes and Ioannis Iossifidis},
  
  doi = {10.48550/arXiv.2304.01355},
  
  year  = {2023},
  
  date = {2023-05-02},
  
  urldate = {2023-05-02},
  
  journal = {arXiv:2304.01355 [cs, math, q-bio]},
  
  keywords = {BCI, Machine Learning, Spike Sorting},
  
  pubstate = {published},
  
  tppubtype = {article}
  
  }
  
  

  Schließen

  • doi:10.48550/arXiv.2304.01355

  Schließen

  2022

  73.

  Grün, Felix; Saif-ur-Rehman, Muhammad; Glasmachers, Tobias; Iossifidis, Ioannis

  Invariance to Quantile Selection in Distributional Continuous Control Artikel

  In: arXiv:2212.14262 [cs.LG], 2022.

  Links | BibTeX | Schlagwörter: Artificial Intelligence (cs.AI), FOS: Computer and information sciences, I.2.6, I.2.8, Machine Learning, Machine Learning (cs.LG)

  @article{grunInvarianceQuantileSelection2022,
  
  title = {Invariance to Quantile Selection in Distributional Continuous Control},
  
  author = {Felix Grün and Muhammad Saif-ur-Rehman and Tobias Glasmachers and Ioannis Iossifidis},
  
  url = {https://arxiv.org/abs/2212.14262},
  
  doi = {10.48550/ARXIV.2212.14262},
  
  year  = {2022},
  
  date = {2022-12-29},
  
  urldate = {2022-12-29},
  
  journal = {arXiv:2212.14262 [cs.LG]},
  
  keywords = {Artificial Intelligence (cs.AI), FOS: Computer and information sciences, I.2.6, I.2.8, Machine Learning, Machine Learning (cs.LG)},
  
  pubstate = {published},
  
  tppubtype = {article}
  
  }
  
  

  Schließen

  • https://arxiv.org/abs/2212.14262
  • doi:10.48550/ARXIV.2212.14262

  Schließen

  72.

  Lehmler, Stephan Johann; Saif-ur-Rehman, Muhammad; Glasmachers, Tobias; Iossifidis, Ioannis

  Deep transfer learning compared to subject-specific models for sEMG decoders Artikel

  In: Journal of Neural Engineering, Bd. 19, Nr. 5, 2022.

  Abstract | Links | BibTeX | Schlagwörter: BCI, Computational Complexity, Deep Transfer-Learning, Machine Learning, transfer learning

  @article{lehmlerTransferLearningPatientSpecific2021bb,
  
  title = {Deep transfer learning compared to subject-specific models for sEMG decoders},
  
  author = {Stephan Johann Lehmler and Muhammad Saif-ur-Rehman and Tobias Glasmachers and Ioannis Iossifidis},
  
  editor = {{IOP Publishing},
  
  url = {https://dx.doi.org/10.1088/1741-2552/ac9860},
  
  doi = {10.1088/1741-2552/ac9860},
  
  year  = {2022},
  
  date = {2022-10-28},
  
  urldate = {2022-10-28},
  
  journal = {Journal of Neural Engineering},
  
  volume = {19},
  
  number = {5},
  
  abstract = {{Objective. Accurate decoding of surface electromyography (sEMG) is pivotal for muscle-to-machine-interfaces and their application e.g. rehabilitation therapy. sEMG signals have high inter-subject variability, due to various factors, including skin thickness, body fat percentage, and electrode placement. Deep learning algorithms require long training time and tend to overfit if only few samples are available. In this study, we aim to investigate methods to calibrate deep learning models to a new user when only a limited amount of training data is available. Approach. Two methods are commonly used in the literature, subject-specific modeling and transfer learning. In this study, we investigate the effectiveness of transfer learning using weight initialization for recalibration of two different pretrained deep learning models on new subjects data and compare their performance to subject-specific models. We evaluate two models on three publicly available databases (non invasive adaptive prosthetics database 2–4) and compare the performance of both calibration schemes in terms of accuracy, required training data, and calibration time. Main results. On average over all settings, our transfer learning approach improves 5%-points on the pretrained models without fine-tuning, and 12%-points on the subject-specific models, while being trained for 22% fewer epochs on average. Our results indicate that transfer learning enables faster learning on fewer training samples than user-specific models. Significance. To the best of our knowledge, this is the first comparison of subject-specific modeling and transfer learning. These approaches are ubiquitously used in the field of sEMG decoding. But the lack of comparative studies until now made it difficult for scientists to assess appropriate calibration schemes. Our results guide engineers evaluating similar use cases.},
  
  keywords = {BCI, Computational Complexity, Deep Transfer-Learning, Machine Learning, transfer learning},
  
  pubstate = {published},
  
  tppubtype = {article}
  
  }
  
  

  Schließen

  {Objective. Accurate decoding of surface electromyography (sEMG) is pivotal for muscle-to-machine-interfaces and their application e.g. rehabilitation therapy. sEMG signals have high inter-subject variability, due to various factors, including skin thickness, body fat percentage, and electrode placement. Deep learning algorithms require long training time and tend to overfit if only few samples are available. In this study, we aim to investigate methods to calibrate deep learning models to a new user when only a limited amount of training data is available. Approach. Two methods are commonly used in the literature, subject-specific modeling and transfer learning. In this study, we investigate the effectiveness of transfer learning using weight initialization for recalibration of two different pretrained deep learning models on new subjects data and compare their performance to subject-specific models. We evaluate two models on three publicly available databases (non invasive adaptive prosthetics database 2–4) and compare the performance of both calibration schemes in terms of accuracy, required training data, and calibration time. Main results. On average over all settings, our transfer learning approach improves 5%-points on the pretrained models without fine-tuning, and 12%-points on the subject-specific models, while being trained for 22% fewer epochs on average. Our results indicate that transfer learning enables faster learning on fewer training samples than user-specific models. Significance. To the best of our knowledge, this is the first comparison of subject-specific modeling and transfer learning. These approaches are ubiquitously used in the field of sEMG decoding. But the lack of comparative studies until now made it difficult for scientists to assess appropriate calibration schemes. Our results guide engineers evaluating similar use cases.

  Schließen

  • https://dx.doi.org/10.1088/1741-2552/ac9860

  Schließen

  71.

  Grün, Felix; Iossifidis, Ioannis

  Exploring Distribution Parameterizations for Distributional Continuous Control Proceedings Article

  In: BC22 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022, BCCN Bernstein Network Computational Network, 2022.

  Links | BibTeX | Schlagwörter: Machine Learning, Reinforcement learning

  @inproceedings{grunExploringDistributionParameterizations2022,
  
  title = {Exploring Distribution Parameterizations for Distributional Continuous Control},
  
  author = {Felix Grün and Ioannis Iossifidis},
  
  doi = {10.12751/nncn.bc2022.112},
  
  year  = {2022},
  
  date = {2022-09-15},
  
  urldate = {2022-09-15},
  
  booktitle = {BC22 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022},
  
  publisher = {BCCN Bernstein Network Computational Network},
  
  keywords = {Machine Learning, Reinforcement learning},
  
  pubstate = {published},
  
  tppubtype = {inproceedings}
  
  }
  
  

  Schließen

  • doi:10.12751/nncn.bc2022.112

  Schließen

  70.

  Lehmler, Stephan Johann; Saif-ur-Rehman, Muhammad; Iossifidis, Ioannis

  Modeling Subject Specfic Surface EMG Features by Means of Deep Learning Proceedings Article

  In: BC22 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022, BCCN Bernstein Network Computational Network, 2022.

  Links | BibTeX | Schlagwörter: BCI, Machine Learning

  @inproceedings{lehmlerModelingSubjectSpecfic2022,
  
  title = {Modeling Subject Specfic Surface EMG Features by Means of Deep Learning},
  
  author = {Stephan Johann Lehmler and Muhammad Saif-ur-Rehman and Ioannis Iossifidis},
  
  doi = {10.12751/nncn.bc2022.309},
  
  year  = {2022},
  
  date = {2022-09-15},
  
  urldate = {2022-09-15},
  
  booktitle = {BC22 : Computational Neuroscience & Neurotechnology Bernstein Conference 2022},
  
  publisher = {BCCN Bernstein Network Computational Network},
  
  keywords = {BCI, Machine Learning},
  
  pubstate = {published},
  
  tppubtype = {inproceedings}
  
  }
  
  

  Schließen

  • doi:10.12751/nncn.bc2022.309

  Schließen

  79 Einträge « ‹ 1 von 8 › »