22 Hits in 6.0 sec

Matching Deformable Objects in Clutter

Luca Cosmo, Emanuele Rodola, Jonathan Masci, Andrea Torsello, Michael M. Bronstein
2016 2016 Fourth International Conference on 3D Vision (3DV)  
Key ingredient to our method is the choice of representation: we formulate the problem in the spectral domain using the functional maps framework, where we seek for the most regular nearly-isometric parts  ...  in the model and the scene that minimize correspondence error.  ...  Assume to be given a set of knowingly similar and dissimilar pairs of points, respectively S and D, and let F Θ (x) be modeled as a deep neural network with trainable parameters Θ.  ... 
doi:10.1109/3dv.2016.10 dblp:conf/3dim/CosmoRMTB16 fatcat:m6bkp7vprbhazoektbx4grnwna

Regularizing Neural Networks via Minimizing Hyperspherical Energy [article]

Rongmei Lin, Weiyang Liu, Zhen Liu, Chen Feng, Zhiding Yu, James M. Rehg, Li Xiong, Le Song
2020 arXiv   pre-print
In this paper, we first study the important role that hyperspherical energy plays in neural network training by analyzing its training dynamics.  ...  To address these problems, we propose the compressive minimum hyperspherical energy (CoMHE) as a more effective regularization for neural networks.  ...  The research is partially supported by NSF BigData program under IIS-1838200, NSF CPS program under CMMI-1932187, and USDOT via C2SMART under 69A3551747124.  ... 
arXiv:1906.04892v2 fatcat:jnbtt6ysvffbxocpikkenf5ype

The Modern Mathematics of Deep Learning [article]

Julius Berner, Philipp Grohs, Gitta Kutyniok, Philipp Petersen
2021 arXiv   pre-print
These questions concern: the outstanding generalization power of overparametrized neural networks, the role of depth in deep architectures, the apparent absence of the curse of dimensionality, the surprisingly  ...  For selected approaches, we describe the main ideas in more detail.  ...  Since the parameters β and γ were introduced, including a batch normalization block also increases the dimension of the optimization problem by two. 6.5 Sparse neural networks and pruning For deep  ... 
arXiv:2105.04026v1 fatcat:lxnfyzr6qfasneo433inpgseia

In search for an alternative to the computer metaphor of the mind and brain [article]

Damian G. Kelty-Stephen, Paul E. Cisek, Benjamin De Bari, James Dixon, Luis H. Favela, Fred Hasselman, Fred Keijzer, Vicente Raja, Jeffrey B. Wagman, Brandon J. Thomas, Madhur Mangalam
2022 arXiv   pre-print
The brain-as-computer metaphor has anchored the professed computational nature of the mind, wresting it down from the intangible logic of Platonic philosophy to a material basis for empirical science.  ...  Despite agreeing about feeling the strain of the strictures of computer metaphors, the authors suggest an exciting diversity of possible metaphoric options for future research into the mind and brain.  ...  form of neural networks.  ... 
arXiv:2206.04603v1 fatcat:qek37aola5bovbsygsat4b5hla

Patterns, predictions, and actions: A story about machine learning [article]

Moritz Hardt, Benjamin Recht
2021 arXiv   pre-print
Starting with the foundations of decision making, we cover representation, optimization, and generalization as the constituents of supervised learning.  ...  Self-contained introductions to causality, the practice of causal inference, sequential decision making, and reinforcement learning equip the reader with concepts and tools to reason about actions and  ...  of neural networks.  ... 
arXiv:2102.05242v2 fatcat:wy47g4fojnfuxngklyewtjtqdi

Multiway Non-rigid Point Cloud Registration via Learned Functional Map Synchronization [article]

Jiahui Huang, Tolga Birdal, Zan Gojcic, Leonidas J. Guibas, Shi-Min Hu
2022 arXiv   pre-print
To maximally benefit from the multi-way information provided by the inferred pairwise deformation fields, we synchronize the pairwise functional maps into a cycle-consistent whole thanks to our novel and  ...  cases in a unified framework and avoid the costly optimization over point-wise permutations by the use of basis function maps.  ...  Dashed rectangles highlight the differences. is an important yet often missing ingredient from many methods.  ... 
arXiv:2111.12878v2 fatcat:i6nqt7f7vfctdi3ow2zjan25ci

Model-based compressive sensing for signal ensembles

Marco F. Duarte, Volkan Cevher, Richard G. Baraniuk
2009 2009 47th Annual Allerton Conference on Communication, Control, and Computing (Allerton)  
The CS literature has focused almost exclusively on problems involving single signals in one or two dimensions. However, many important applications involve distributed networks or arrays of sensors.  ...  The Kronecker product formulation in the sparsity iii and measurement settings enables the derivation of analytical bounds for transform coding compression of signal ensembles and multidimensional signals  ...  Such a structure has applications not only in sensors and surveillance networks, but also in the localization of action potentials in multineuron recordings, where the neural activity or spikes are recorded  ... 
doi:10.1109/allerton.2009.5394807 fatcat:73c2lpncizdxxd5jowblg3dwbm

Special issue on information reuse and integration

2007 IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics)  
Based on our results, we recommend using the random forest ensemble learning technique for building classification models from software measurement data, regardless of the quality and class distribution  ...  Low quality or noisy data, which typically consists of erroneous values for both dependent and independent variables, has been demonstrated to have a significantly negative impact on the classification  ...  Acknowledgement We wish to thank the Overview Editor, Dr. Maria Ganzha, for her consideration of this paper.  ... 
doi:10.1109/tsmcb.2007.912701 fatcat:xvhlaf4m3vhcdb2cicqtfrug6m

Reflections on the four facets of symmetry: how physics exemplifies rational thinking

Amaury Mouchet
2013 The European Physical Journal H  
In contemporary theoretical physics, the powerful notion of symmetry stands for a web of intricate meanings among which I identify four clusters associated with the notion of transformation, comprehension  ...  This decomposition allows us to examine closely the multiple different roles symmetry plays in many places in physics.  ...  Villain for precious discussions that allowed to selectively improve the formulation of some ideas presented here. I am also very grateful towards N. Mohammedi for his careful corrections.  ... 
doi:10.1140/epjh/e2013-40018-4 fatcat:u2cexlbcv5hdbjw64lkmtexfoq

A Unified Approach to Sparse Signal Processing [article]

F. Marvasti, A. Amini, F. Haddadi, M. Soltanolkotabi, B. H. Khalaj, A. Aldroubi, S. Holm, S. Sanei, J. Chambers
2009 arXiv   pre-print
The notions of sparse array beamforming and sparse sensor networks are also introduced.  ...  The methods of Prony, Pisarenko, and MUSIC are next discussed for sparse frequency domain representations.  ...  They proposed a solution based on a recurrent artificial neural network for separation of the sources.  ... 
arXiv:0902.1853v1 fatcat:lxd3crtgd5fl5ofshryzm5pkfi

Automating Ambiguity: Challenges and Pitfalls of Artificial Intelligence [article]

Abeba Birhane
2022 arXiv   pre-print
for approaching challenges, failures and problems surrounding ML systems as well as alternative ways forward.  ...  They remain opaque and unreliable, and fail to consider societal and structural oppressive systems, disproportionately negatively impacting those at the margins of society while benefiting the most powerful  ...  Feed a neural network labelled images of faces and it will learn to discern faces from not-faces.  ... 
arXiv:2206.04179v1 fatcat:qqln6jaiznctzcshmnsvvv73y4

Crawler [chapter]

Kenneth A. Ross, Christian S. Jensen, Richard Snodgrass, Curtis E. Dyreson, Christian S. Jensen, Richard Snodgrass, Spiros Skiadopoulos, Cristina Sirangelo, Mary Lynette Larsgaard, Gösta Grahne, Daniel Kifer, Hans-Arno Jacobsen (+106 others)
2009 Encyclopedia of Database Systems  
For such workloads, improving the locality of data-intensive operations can have a direct impact on the system's overall performance.  ...  As a result, it is common (at the time of writing) for data accesses to RAM to require several hundred CPU cycles to resolve.  ...  model of neural networks.  ... 
doi:10.1007/978-0-387-39940-9_2315 fatcat:x4qspjdytvhvroc7h753dihp7u

The Calabi-Yau Landscape: from Geometry, to Physics, to Machine-Learning [article]

Yang-Hui He
2020 arXiv   pre-print
to ArXiv for comments and suggestions.  ...  Aimed at the beginning research student and using Calabi-Yau spaces as an exciting play-ground, we intend to teach some mathematics to the budding physicist, some physics to the budding mathematician,  ...  likes of a neural network or SVM, and how.  ... 
arXiv:1812.02893v2 fatcat:fvjjtjo2tngjzl6ubuv42f56va

Bayesian learning for nonlinear system identification

Wei Pan, Guy-Bart Stan
In particular, this thesis considers sparse modelling and estimation for a selection of nonlinear dynamical systems classes.  ...  Although, a large amount of data are being collected on a daily basis, very few methods allow the automatic creation from these data of nonlinear dynamical models for understanding and (re-)design/control  ...  networks and recurrent neural network.  ... 
doi:10.25560/68510 fatcat:ko656tegpvgebjnrrqlgostkhu

Quantum machine learning for chemistry and physics

Manas Sajjan, Junxu Li, Raja Selvarajan, Shree Hari Sureshbabu, Sumit Suresh Kale, Rishabh Gupta, Vinit Singh, Sabre Kais
Machine learning (ML) has emerged as a formidable force for identifying hidden but pertinent patterns within a given data set with the objective of subsequent generation of automated predictive behavior  ...  performance of photovoltaics, electronic structure calculations of ground and excited states of correlated matter, computation of force-fields and potential energy surfaces informing chemical reaction dynamics  ...  Acknowledgements The authors acknowledge funding by the U.S. Department  ... 
doi:10.1039/d2cs00203e pmid:35849066 fatcat:iudcpxm6crhirjsswfqsjxpowm
« Previous Showing results 1 — 15 out of 22 results