Filters








7,913 Hits in 7.8 sec

Analytic Manifold Learning: Unifying and Evaluating Representations for Continuous Control [article]

Rika Antonova, Maksim Maydanskiy, Danica Kragic, Sam Devlin, Katja Hofmann
2020 arXiv   pre-print
Our second contribution is a unifying mathematical formulation for learning latent relations.  ...  We present mathematical properties, concrete algorithms for implementation and experimental validation of successful learning and transfer of latent relations.  ...  Evaluation Suite for Unsupervised Learning for Continuous Control Reinforcement learning (RL) has shown strong progress recently [4] , and RL for continuous control is particularly promising for robotics  ... 
arXiv:2006.08718v2 fatcat:ag2hswmjbzfrpfnhnf7h24whqe

Variational Autoencoders for Learning Nonlinear Dynamics of Physical Systems [article]

Ryan Lopez, Paul J. Atzberger
2021 arXiv   pre-print
We develop ways to incorporate geometric and topological priors through general manifold latent space representations.  ...  We investigate the performance of our methods for learning low dimensional representations for the nonlinear Burgers equation and constrained mechanical systems.  ...  Authors also acknowledge UCSB Center for Scientific Computing NSF MR-SEC (DMR1121053) and UCSB MRL NSF CNS-1725797. P.J.A. would also like to acknowledge a hardware grant from Nvidia.  ... 
arXiv:2012.03448v2 fatcat:atf6ucqe4rd6rklsmyi4kep4gi

Network Representation Learning: From Traditional Feature Learning to Deep Learning

Ke Sun, Lei Wang, Bo Xu, Wenhong Zhao, Shyh Wei Teng, Feng Xia
2020 IEEE Access  
MANIFOLD LEARNING Manifold learning methods focus on preserving local similarity among data when the new representations are learned.  ...  These vertex vectors of edges stay in a continued space and we have u + l ≈ v , (26) where u and v denote the representations of vertices, and l is the edge representation derived from label set l.  ... 
doi:10.1109/access.2020.3037118 fatcat:kca6htfarjdjpmtwcvbsppfzui

Gaussians on Riemannian Manifolds: Applications for Robot Learning and Adaptive Control [article]

Sylvain Calinon
2020 arXiv   pre-print
This article presents an overview of robot learning and adaptive control applications that can benefit from a joint use of Riemannian geometry and probabilistic representations.  ...  A varied range of techniques employing Gaussian distributions on Riemannian manifolds is then introduced, including clustering, regression, information fusion, planning and control problems.  ...  for statistical learning, dynamical systems, optimal control and Riemannian geometry.  ... 
arXiv:1909.05946v4 fatcat:ojaty7ptljdblbrt7p6jwiom3a

Variational Autoencoders for Learning Nonlinear Dynamics of PDEs and Reductions

Ryan Lopez, Paul J. Atzberger
2021 AAAI Spring Symposia  
We develop ways to incorporate geometric and topological priors through general manifold latent space representations.  ...  We investigate the performance of our methods for learning low dimensional representations for the nonlinear Burgers equation and constrained mechanical systems.  ...  Authors also acknowledge UCSB Center for Scientific Computing NSF MR-SEC (DMR1121053) and UCSB MRL NSF CNS-1725797. P.J.A. would also like to acknowledge a hardware grant from Nvidia.  ... 
dblp:conf/aaaiss/LopezA21 fatcat:lcqnltagwvaitey5osscnfwhbq

GD-VAEs: Geometric Dynamic Variational Autoencoders for Learning Nonlinear Dynamics and Dimension Reductions [article]

Ryan Lopez, Paul J. Atzberger
2022 arXiv   pre-print
GD-VAEs provide methods for obtaining representations for use in learning tasks involving dynamics.  ...  We develop approaches for learning nonlinear state space models of the dynamics for general manifold latent spaces using training strategies related to Variational Autoencoders (VAEs).  ...  Authors also acknowledge UCSB Center for Scientific Computing NSF MRSEC (DMR1121053) and UCSB MRL NSF CNS-1725797. P.J.A. would also like to acknowledge a hardware grant from Nvidia.  ... 
arXiv:2206.05183v1 fatcat:l3kppxktt5b57jjpwg7vzs354a

Network representation learning systematic review: Ancestors and current development state

Amina Amara, Mohamed Ali Hadj Taieb, Mohamed Ben Aouicha
2021 Machine Learning with Applications  
Most commonly used downstream tasks to evaluate embeddings, their evaluation metrics and popular datasets are highlighted. Finally, we present the open-source libraries for network embedding.  ...  Thus, we introduce a brief history of representation learning and word representation learning ancestor of network embedding.  ...  A natural coordinate system is provided by this manifold to the representation being learned. Proposed approaches for representation learning cover a wide range of applications.  ... 
doi:10.1016/j.mlwa.2021.100130 fatcat:axhg2gxkzfds3icebro6hlman4

Survey on lie group machine learning

Mei Lu, Fanzhang Li
2020 Big Data Mining and Analytics  
, Finsler geometric learning, homology boundary learning, category representation learning, and neuromorphic synergy learning.  ...  It will enable researchers to comprehensively understand the state of the field, identify the most appropriate tools for particular applications, and identify directions for future research.  ...  Hence, the fiber bundle can be used as a tool for the further analytical processing of the manifold structure and its tangent bundle.  ... 
doi:10.26599/bdma.2020.9020011 fatcat:ki4l7b5o6ncltfgcu462flxsiq

A unified probabilistic framework for robust manifold learning and embedding

Qi Mao, Li Wang, Ivor W. Tsang
2016 Machine Learning  
graph and the positive weights stand for the new similarity.  ...  To achieve this goal, we propose a unified probabilistic framework that directly models the posterior distribution of data points in an embedding space so as to suppress data noise and reveal the smooth  ...  Acknowledgements This project is partially supported by the ARC Future Fellowship FT130100746 and ARC grant LP150100671.  ... 
doi:10.1007/s10994-016-5602-8 fatcat:fv62enupyze2rabxea5htwteq4

Graph representation learning: a survey

Fenxiao Chen, Yun-Cheng Wang, Bin Wang, C.-C. Jay Kuo
2020 APSIPA Transactions on Signal and Information Processing  
Research on graph representation learning has received great attention in recent years since most data in real-world applications come in the form of graphs.  ...  Then, we evaluate several stat-of-the-art methods against small and large data sets and compare their performance. Finally, potential applications and future directions are presented.  ...  Then, non-linear dimensionality reduction graph representation learning: a survey (NLDR) [42] can be used for manifold learning. The objective is to learn the nonlinear topology automatically.  ... 
doi:10.1017/atsip.2020.13 fatcat:lirq3kp25jfilgkf66u2rlkhky

Representation Learning for Fine-Grained Change Detection

Niall O'Mahony, Sean Campbell, Lenka Krpalkova, Anderson Carvalho, Joseph Walsh, Daniel Riordan
2021 Sensors  
This review focuses on the state-of-the-art methods, applications, and challenges of representation learning for fine-grained change detection.  ...  Our research focuses on methods of harnessing the latent metric space of representation learning techniques as an interim output for hybrid human-machine intelligence.  ...  Acknowledgments: The authors wish to acknowledge the DJEI/DES/SFI/HEA Irish Centre for High-End Computing (ICHEC) for the provision of computational facilities and support.  ... 
doi:10.3390/s21134486 pmid:34209075 fatcat:2shbcyvutvfbhhqrsmjipjdrzi

Deformed Kernel Based Extreme Learning Machine

Chen Zhang, Xiong Shi Xia, Bing Liu
2013 Journal of Computers  
is built with both labeled and unlabeled samples.  ...  The extreme learning machine (ELM) is a newly emerging supervised learning method.  ...  Based on the deformed kernel, we propose a deformed kernel-based extreme learning machine (DKELM) to provide a unified solution for regression, binary, and multiclass classifications (like ELM).  ... 
doi:10.4304/jcp.8.6.1602-1609 fatcat:kwpr3yvtvvbajmv3tdnmfn3ace

Machine Learning in Chemical Engineering : A Perspective

Artur M. Schweidtmann, Erik Esche, Asja Fischer, Marius Kloft, Jens-Uwe Repke, Sebastian Sager, Alexander Mitsos
2021 Chemie - Ingenieur - Technik (2021). doi:10.1002/cite.202100083  
representation, (4) heterogeneity of data, (5) safety and trust in ML applications, and (6) creativity.  ...  We identify six challenges that will open new methods for CE and formulate new types of problems for ML: (1) optimal decision making, (2) introducing and enforcing physics in ML, (3) information and knowledge  ...  The authors gratefully acknowledge the DFG for establishing the Priority Programme SPP 2331 ''Machine learning in chemical engineering.  ... 
doi:10.18154/rwth-2021-09826 fatcat:7tlvcx22urd27fpjfbw4iqhr7a

Probabilistic Dimensionality Reduction via Structure Learning [article]

Li Wang
2016 arXiv   pre-print
We develop a new method to learn the embedding points that form a spanning tree, which is further extended to obtain a discriminative and compact feature representation for clustering problems.  ...  The formulation of the new model can be equivalently interpreted as two coupled learning problem, i.e., structure learning and the learning of projection matrix.  ...  Except for the special case, problem (30) is able to achieve discriminative and compact feature representation for dimensionality reduction since clustering objective and DRTree are optimized in a unified  ... 
arXiv:1610.04929v1 fatcat:vfa2vzayqve3fltm6srmytzrvy

Structure preserving deep learning [article]

Elena Celledoni, Matthias J. Ehrhardt, Christian Etmann, Robert I McLachlan, Brynjulf Owren, Carola-Bibiane Schönlieb, Ferdia Sherry
2020 arXiv   pre-print
between computational effort, amount of data and model complexity is required to successfully design a deep learning approach for a given problem.  ...  new algorithmic frameworks based on conformal Hamiltonian systems and Riemannian manifolds to solve the optimisation problems have been proposed.  ...  The authors would like to thank the Isaac Newton Institute for Mathematical Sciences, Cambridge, for support and hospitality during the programmes Variational methods and effective algorithms for imaging  ... 
arXiv:2006.03364v1 fatcat:dy5t5w4gsfeavl72e3oqllnlqe
« Previous Showing results 1 — 15 out of 7,913 results