Filters








78,547 Hits in 4.6 sec

Learning to Fly by MySelf: A Self-Supervised CNN-Based Approach for Autonomous Navigation

Alexandros Kouris, Christos-Savvas Bouganis
2018 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
distance to its closest obstacle towards multiple directions.  ...  In this work, we introduce a self-supervised CNN-based approach for indoor robot navigation.  ...  Having used pairs of both sensor technologies towards each direction, the recorded distance data from each pair are used to automatically annotate the corresponding input images of each trajectory by being  ... 
doi:10.1109/iros.2018.8594204 dblp:conf/iros/KourisB18 fatcat:hbnttqeklzfitamruim2ax5wgm

Trajectory-based Algorithm Selection with Warm-starting [article]

Anja Jankovic, Diederick Vermetten, Ana Kostovska, Jacob de Nobel, Tome Eftimov, Carola Doerr
2022 arXiv   pre-print
Features computed in this manner are used to train algorithm performance regression models, upon which a per-run algorithm selector is then built.  ...  In this new context, we show promising performance of the trajectory-based per-run algorithm selection with warm-starting.  ...  The trajectory-based algorithm selection can hence be considered an extension of the per-instance algorithm selection towards a per-run algorithm selection. III.  ... 
arXiv:2204.06397v1 fatcat:qf7ywzicxfg4pkn6otevzok3gu

Modeling and prediction of clinical symptom trajectories in Alzheimer's disease using longitudinal data

Nikhil Bhagwat, Joseph D. Viviano, Aristotle N. Voineskos, M. Mallar Chakravarty, Saad Jbabdi
2018 PLoS Computational Biology  
predicting these trajectories using data from both baseline and a follow-up visits within one year.  ...  six years based on a hierarchical clustering approach.  ...  based on the data that was used.  ... 
doi:10.1371/journal.pcbi.1006376 pmid:30216352 pmcid:PMC6157905 fatcat:eiflnxqj5fhsrh6vrxx3rsrt4a

Per-run Algorithm Selection with Warm-starting using Trajectory-based Features [article]

Ana Kostovska, Anja Jankovic, Diederick Vermetten, Jacob de Nobel, Hao Wang, Tome Eftimov, Carola Doerr
2022 arXiv   pre-print
The selection is classically done offline, using openly available information about the problem instance or features that are extracted from the instance during a dedicated feature extraction step.  ...  We also compare two different feature extraction principles, based on exploratory landscape analysis and time series analysis of the internal state variables of the CMA-ES, respectively.  ...  We create a mapping between the input feature data, which can be one of the following: the trajectory-based representation with 38 ELA features per run (ELAbased AS) the trajectory-based representation  ... 
arXiv:2204.09483v1 fatcat:rpxh5ibl6rebdngmdz7tmz6lme

Decoding Imagined 3D Arm Movement Trajectories From EEG to Control Two Virtual Arms—A Pilot Study

Attila Korik, Ronen Sosnik, Nazmul Siddique, Damien Coyle
2019 Frontiers in Neurorobotics  
The 3D trajectory of imagined arm movements was decoded from power spectral density of mu, low beta, high beta, and low gamma EEG oscillations using multiple linear regression.  ...  Target classification accuracy using predicted trajectories of the virtual arms was computed and compared with results of a filter-bank common spatial patterns (FBCSP) based multi-class classification  ...  AUTHOR CONTRIBUTIONS AK carried out the research, data collection, computation and data analysis, and prepared the figures and draft of the manuscript. RS reviewed and revised the manuscript.  ... 
doi:10.3389/fnbot.2019.00094 pmid:31798438 pmcid:PMC6868122 fatcat:nn6r5zytivdtphnim5q3l6furi

Mapping between acoustic and articulatory gestures

G. Ananthakrishnan, Olov Engwall
2011 Speech Communication  
The Acoustic-to-Articulatory Inversion is performed using a GMM-based regression and the results are at par with state-of-the-art frame-based methods with dynamical constraints (with an average error of  ...  A definition for these gestures along with a method to segment the measured articulatory trajectories and the acoustic waveform into gestures is suggested.  ...  The MMSE and MLTE methods are the traditional Frame Based (FB) methods, without and with dynamic features respectively, while the Gesture based method uses the same GMMR regression, but has gesture based  ... 
doi:10.1016/j.specom.2011.01.009 fatcat:ulav2err4jewtfl27gjp6mect4

Towards Visual Ego-motion Learning in Robots [article]

Sudeep Pillai, John J. Leonard
2017 arXiv   pre-print
We envision robots to be able to learn and perform these tasks, in a minimally supervised setting, as they gain more experience.  ...  Many model-based Visual Odometry (VO) algorithms have been proposed in the past decade, often restricted to the type of camera optics, or the underlying motion manifold observed.  ...  This allows us to fine-tune the regressor to predict valid estimates that integrate towards accurate longterm ego-motion trajectories.  ... 
arXiv:1705.10279v1 fatcat:n6nugql2j5fitpl2qgxjthrxe4

Understanding Spending Behavior: Recurrent Neural Network Explanation and Interpretation [article]

Charl Maree, Christian W. Omlin
2021 arXiv   pre-print
We show that our linear regression coefficients have not only learned the rules used to recreate the features, but have also learned the relationships that were not directly evident in the raw data.  ...  For the explanation, we use a linear regression model to reconstruct the features in the state space with high fidelity.  ...  We show that our linear regression coefficients have not only learned the rules used to recreate the features, but have also learned the relationships that were not directly evident in the raw data.  ... 
arXiv:2109.11871v2 fatcat:wnhgxxw27nfa3fcku3te3boopm

Incorporating Human Domain Knowledge into Large Scale Cost Function Learning [article]

Markus Wulfmeier, Dushyant Rao, Ingmar Posner
2016 arXiv   pre-print
Our work achieves this by pretraining a model to regress to a manual cost function and refining it based on Maximum Entropy Deep Inverse Reinforcement Learning.  ...  data from human drivers.  ...  We will show quantitatively that regression based pretraining improves prediction performance as well as classification performance for traversable terrain.  ... 
arXiv:1612.04318v1 fatcat:abb7xedtazcblpbpetxpnd2nii

Automated Video Analysis of Handwashing Behavior as a Potential Marker of Cognitive Health in Older Adults

Ahmed Ashraf, Babak Taati
2016 IEEE journal of biomedical and health informatics  
To assess multivariate performance, a regression model was trained using leave-one-subject-out cross validation.  ...  The predicted MMSE scores using the regression model based on uncollapsed representation had a mild correlation (R = 0.663, p < 0.001), while the model based on collapsed features had a moderate to strong  ... 
doi:10.1109/jbhi.2015.2413358 pmid:25794404 fatcat:7nr2fa5vtbcpzbzuricarm5bhq

Towards Uncertainty Quantification for Electrode Bending Prediction in Stereotactic Neurosurgery

Alejandro Granados, Oeslle Lucena, Vejay Vakharia, Anna Miserocchi, Andrew W. McEvoy, Sjoerd B. Vos, Roman Rodionov, John S. Duncan, Rachel Sparks, Sebastien Ourselin
2020 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI)  
We compute image features of 23 stereoelectroencephalography cases (241 electrodes) and use them as inputs to a neural network to regress electrode local displacement.  ...  Although mechanical models and data-driven approaches have been proposed for trajectory prediction, they lack to report uncertainty of the predictions.  ...  Ethical Approval All data were evaluated retrospectively.  ... 
doi:10.1109/isbi45749.2020.9098730 dblp:conf/isbi/GranadosLVMMVRD20 fatcat:az44igmorfhcrexvoyryoucvxe

Online Multi-Object Tracking Based on Feature Representation and Bayesian Filtering within a Deep Learning Architecture

Jun Xiang, Guoshuai Zhang, Jianhua Hou
2019 IEEE Access  
In detection-based multi-object tracking (MOT), one challenging problem is to design a robust affinity model for data association.  ...  INDEX TERMS Multi-object tracking, deep learning, data association, trajectory reconstruction.  ...  In their work, a RNN-based architecture is used for Bayesian state estimation, i.e. trajectory estimation, and a LSTM-based model is designed for data association.  ... 
doi:10.1109/access.2019.2901520 fatcat:yabmic6svnbx5ljbfaaeubt4zi

Patient-specific prediction of SEEG electrode bending for stereotactic neurosurgical planning

Alejandro Granados, Yuxuan Han, Oeslle Lucena, Vejay Vakharia, Roman Rodionov, Sjoerd B. Vos, Anna Miserocchi, Andrew W. McEvoy, John S. Duncan, Rachel Sparks, Sébastien Ourselin
2021 International Journal of Computer Assisted Radiology and Surgery  
Results mage-based models outperformed features-based models for all groups, and models that predicted $$\mathbf{lu} $$ lu performed better than for $$\hat{\mathbf{eb }}$$ eb ^ .  ...  Methods We transform electrodes of 86 cases into a common space and compare features-based and image-based neural networks on their ability to regress local displacement ($$\mathbf{lu} $$ lu ) or electrode  ...  The aim of this work is to: 1) assess two data-driven approaches for predicting implanted electrode trajectories using a total of 96 handcrafted features or using electrode direction and a 3D image and  ... 
doi:10.1007/s11548-021-02347-8 pmid:33761063 fatcat:zxeyarupbrbcrlov2unodowhme

Learning stable pushing locations

Tucker Hermans, Fuxin Li, James M. Rehg, Aaron F. Bobick
2013 2013 IEEE Third Joint International Conference on Development and Learning and Epigenetic Robotics (ICDL)  
The robot performs push experiments at many contact locations on multiple objects and records local and global shape features at each point of contact.  ...  The robot then learns a regression function in order to predict push effectiveness as a function of object shape.  ...  EXPERIMENTAL PROCEDURE We collected all pushing data using a Willow Garage PR2 robot. We performed all experiments using common household objects in the Georgia Tech Aware Home.  ... 
doi:10.1109/devlrn.2013.6652539 dblp:conf/icdl-epirob/HermansLRB13 fatcat:6mmcfn5jzvgyjcgmw4wn52vkve

Motion generation of robotic surgical tasks: Learning from expert demonstrations

Carol E Reiley, Erion Plaku, Gregory D Hager
2010 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology  
Gaussian Mixture Regression (GMR) is then used to extract a smooth reference trajectory to reproduce a trajectory of the task.  ...  Results suggest that this paper presents a means to (i) extract important features of the task, (ii) create a metric to evaluate robot imitative performance (iii) generate smoother trajectories for reproduction  ...  Finally we are able to generate trajectories to perform gestures similar to the training examples using Gaussian Mixture Regression (GMR) which constrains smoothness.  ... 
doi:10.1109/iembs.2010.5627594 pmid:21096982 fatcat:ok26vsblaraztiybne7a2zfi2q
« Previous Showing results 1 — 15 out of 78,547 results