A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf
.
Filters
An interior-point stochastic approximation method and an L1-regularized delta rule
2008
Neural Information Processing Systems
We establish the stability of a stochastic interior-point approximation method both analytically and empirically, and demonstrate its utility by deriving an on-line learning algorithm that also performs ...
We propose that interior-point methods are a natural solution. ...
Thanks to Ewout van den Berg, Matt Hoffman and Firas Hamze. ...
dblp:conf/nips/CarbonettoSF08
fatcat:fj3i7u3cp5antkibc3kqoac4ly
PySDTest: a Python Package for Stochastic Dominance Tests
[article]
2023
arXiv
pre-print
We briefly give an overview of the concepts of stochastic dominance and testing methods. We then provide a practical guidance for using PySDTest. ...
PySDTest provides several options to compute the critical values including bootstrap, subsampling, and numerical delta methods. ...
() #------------------------------------------- Step size ϵ N for numerical approximation Numerical delta method (NDM) float form Type of functional (KS, L1, and L2) Numerical delta method (NDM) str ...
arXiv:2307.10694v1
fatcat:d46ecso6afdtrgfiisxyp6heyi
Inverse Path Tracing for Joint Material and Lighting Estimation
[article]
2019
arXiv
pre-print
This enables joint optimization for physically correct light transport and material models using a tailored stochastic gradient descent. ...
The key contribution of this work is an accurate and simultaneous retrieval of light sources and physically based material properties (e.g., diffuse reflectance, specular reflectance, roughness, etc.) ...
We would also like to thank Angela Dai for the video voice over and Abhimitra Meka for the LIME comparison. ...
arXiv:1903.07145v1
fatcat:exp5b2tqo5dadpwnu6duoooocy
14th International Symposium on Mathematical Programming
1990
Mathematical programming
It is shown that for the success of the variant dom must ful ll a regularity property and that the choice of the normal vectors must meet some demands.Both requirements are ful lled if dom is polyhedral ...
If we use a decomposition approach in order to solve a minimization problem we often get an objective function in such a w a y that its domain dom 6 = n is not given explicitely to us. ...
interior point methods. ...
doi:10.1007/bf01580875
fatcat:3jtclwmntzgjxkqs5uecombdaa
Challenges in the Application of Mathematical Programming in the Enterprise-wide Optimization of Process Industries
2014
Теоретические основы химической технологии
Barton Solving L1-CTA in 3D tables by an interior-point method for block-angular problems Jordi Cuesta, Jordi Castro 2 -Optimal Data-Independent Noise for Differential Privacy Josep Domingo-Ferrer, Jordi ...
windows by an interior point branch-price-and-cut framework Pedro Munari, Jacek Gondzio
-A Parallel Algorithm for Vehicle Routing Problem on GPUs ErdenerÖzçetin, Gurkan Ozturk
Paper added to session ...
An application to the U.S. financial sector using stochastic frontier models
Hanns de la Fuente Paper added to session
-Efficiency evaluation and analysis of Third Party Logistics in Brazil Mariana ...
doi:10.7868/s0040357114050054
fatcat:kli7aeuyxbaplfhup2t6nmuyxq
Challenges in the application of mathematical programming in the enterprise-wide optimization of process industries
2014
Theoretical foundations of chemical engineering
Barton Solving L1-CTA in 3D tables by an interior-point method for block-angular problems Jordi Cuesta, Jordi Castro 2 -Optimal Data-Independent Noise for Differential Privacy Josep Domingo-Ferrer, Jordi ...
windows by an interior point branch-price-and-cut framework Pedro Munari, Jacek Gondzio
-A Parallel Algorithm for Vehicle Routing Problem on GPUs ErdenerÖzçetin, Gurkan Ozturk
Paper added to session ...
An application to the U.S. financial sector using stochastic frontier models
Hanns de la Fuente Paper added to session
-Efficiency evaluation and analysis of Third Party Logistics in Brazil Mariana ...
doi:10.1134/s0040579514050182
fatcat:3ra5yqooyzgmroo5qccbnauftm
Stochastic Collapsed Variational Bayesian Inference for Latent Dirichlet Allocation
[article]
2013
arXiv
pre-print
We propose a stochastic algorithm for collapsed variational Bayesian inference for LDA, which is simpler and more efficient than the state of the art method. ...
In experiments on large-scale text corpora, the algorithm was found to converge faster and often to a better solution than the previous method. ...
and convergence of a stochastic approximation algorithm. ...
arXiv:1305.2452v1
fatcat:vcl23akbfnerbf5z73afg2iby4
Locally adaptive fitting of semiparametric models to nonstationary time series
2001
Stochastic Processes and their Applications
This method is fully automatic and adapts to di erent smoothness classes. It is shown that usual rates of convergence in Besov smoothness classes are attained up to a logarithmic factor. ...
Whereas the mean function is estimated by a usual kernel estimator, each component of Â(·) is estimated by a nonlinear wavelet method. ...
(iii) Assume in the following that both k;inf and˜ k are interior points. ...
doi:10.1016/s0304-4149(00)00060-0
fatcat:7ypi5rxkirey5lahuu3jtsef6u
Tests for almost stochastic dominance
[article]
2024
arXiv
pre-print
As an application, we develop consistent bootstrap testing procedures for almost stochastic dominance. The performance of the tests is checked via simulations and the analysis of real data. ...
We introduce a 2-dimensional stochastic dominance (2DSD) index to characterize both strict and almost stochastic dominance. ...
Acknowledgements We are very grateful to the Editor, Associate Editor and two reviewers for all the comments on the first version of the paper. ...
arXiv:2403.15258v1
fatcat:lfy576jb2rbblpqw72qnmerlk4
Inverse-Dirichlet Weighting Enables Reliable Training of Physics Informed Neural Networks
[article]
2021
arXiv
pre-print
Their training amounts to solving an optimization problem over a weighted sum of data-fidelity and equation-fidelity objectives. ...
We explain the training pathology arising from this and propose a simple yet effective inverse-Dirichlet weighting strategy to alleviate the issue. ...
ACKNOWLEDGMENTS This work was supported by the German Research Foundation (DFG) -EXC-2068, Cluster of Excellence "Physics of Life", and by the Center for Scalable Data Analytics and Artificial Intelligence ...
arXiv:2107.00940v1
fatcat:tucharqqmve6hiz6625bswhg4m
A δ
2022
ACM Transactions on Graphics
We describe when such approximation rules are first-order correct, and show that this correctness criterion applies to a relatively broad class of functions. ...
Our compiler outputs gradient programs in TensorFlow, PyTorch (for quick prototypes) and Halide with an optional auto-scheduler (for efficiency). ...
Therefore in this section, we compare our method with finite difference and its stochastic variant SPSA [Spall 1992 ]. ...
doi:10.1145/3528223.3530125
fatcat:y6vxpv3trnffjcgvjn6pe4pwhu
A Deep Fourier Residual Method for solving PDEs using Neural Networks
[article]
2022
arXiv
pre-print
The resulting Deep Fourier-based Residual (DFR) method efficiently and accurately approximate solutions to PDEs. ...
This is particularly useful when solutions lack H^2 regularity and methods involving strong formulations of the PDE fail. ...
200 equispaced integration points in the collocation method with a mid-point integration rule. ...
arXiv:2210.14129v1
fatcat:6lt6ps7wyngorddfjnsgjpc7om
Neural Networks are Convex Regularizers: Exact Polynomial-time Convex Optimization Formulations for Two-layer Networks
[article]
2020
arXiv
pre-print
Our theory utilizes semi-infinite duality and minimum norm regularization. We show that ReLU networks trained with standard weight decay are equivalent to block ℓ_1 penalized convex models. ...
exact representations of training two-layer neural networks with rectified linear units (ReLUs) in terms of a single convex program with number of variables polynomial in the number of training samples and ...
Acknowledgements This work was supported in part by the National Science Foundation under grant IIS-1838179 and Stanford SystemX Alliance. ...
arXiv:2002.10553v2
fatcat:4jcbffe2p5bqbmeq4in5baikju
Online Dynamics Learning for Predictive Control with an Application to Aerial Robots
[article]
2022
arXiv
pre-print
In this offline setting, training data is first collected and a prediction model is learned through an elaborated training procedure. ...
To improve the adaptiveness of the model and the controller, we propose an online dynamics learning framework that continually improves the accuracy of the dynamic model during deployment. ...
The authors would also like to thank all reviewers and the area chair for their reviews and comments. ...
arXiv:2207.09344v2
fatcat:hdxjgafnurcgxoiderpcv4riy4
Learning Positive Functions with Pseudo Mirror Descent
2019
Neural Information Processing Systems
The algorithm guarantees positivity by performing mirror descent with an appropriately selected Bregman divergence, and a pseudo-gradient is adopted to speed up the gradient evaluation procedure in practice ...
The nonparametric learning of positive-valued functions appears widely in machine learning, especially in the context of estimating intensity functions of point processes. ...
Consider an iterative algorithm initialized at x (0) and with intermediate updates x (1) , . . . , x (k) , where each x (k) is generated from some given rule r(x (k−1) , g (k) ) with a random direction ...
dblp:conf/nips/YangWKH19
fatcat:24wunft5bzejheobtc7f74e73u
« Previous
Showing results 1 — 15 out of 391 results