Filters








11,797 Hits in 6.8 sec

Learning Feature Embeddings for Discriminant Model based Tracking [article]

Linyu Zheng, Ming Tang, Yingying Chen, Jinqiao Wang, Hanqing Lu
<span title="2020-09-06">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our method, called DCFST, integrates the solver of a discriminant model that is differentiable and has a closed-form solution into convolutional neural networks.  ...  for online discriminative tracking.  ...  ., learning feature embeddings with Differentiable and Closed-Form Solver for Tracking, is efficient and generalizes well to class-agnostic target objects in online tracking, thus achieves state-of-the-art  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1906.10414v2">arXiv:1906.10414v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/kr7vtckz5rcebe6a6yqjgmq73y">fatcat:kr7vtckz5rcebe6a6yqjgmq73y</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201119010126/https://arxiv.org/pdf/1906.10414v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/3c/a1/3ca19da51915b31641340808186272e75485806c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1906.10414v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Learning to Optimize Non-Rigid Tracking [article]

Yang Li, Aljaž Božič, Tianwei Zhang, Yanli Ji, Tatsuya Harada, Matthias Nießner
<span title="2020-03-27">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
One of the widespread solutions for non-rigid tracking has a nested-loop structure: with Gauss-Newton to minimize a tracking objective in the outer loop, and Preconditioned Conjugate Gradient (PCG) to  ...  First, we upgrade the tracking objective by integrating an alignment data term on deep features which are learned end-to-end through CNN.  ...  The experimental results show that the learned non-rigid feature significantly improves the convergence of Gauss-Newton solver for the frame-frame non-rigid tracking.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.12230v1">arXiv:2003.12230v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/n4r25rvhzrc2fahx4fve54pole">fatcat:n4r25rvhzrc2fahx4fve54pole</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200331000826/https://arxiv.org/pdf/2003.12230v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.12230v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Neural Differential Equations as a Basis for Scientific Machine Learning (SciML) [article]

Christopher Rackauckas
<span title="2020-08-03">2020</span> <i title="figshare"> figshare.com </i> &nbsp;
Additionally, deep learning embedded within backwards stochastic differential equations has been shown to be an effective tool for solving high-dimensional partial differential equations, like the Hamilton-Jacobian-Bellman  ...  Scientific Machine Learning (SciML) is an emerging discipline which merges the mechanistic models of science and engineering with non-mechanistic machine learning models to solve problems which were previously  ...   Instead of picking a form for ′ ′(the current method), replace it with a neural network and learn it from small scale simulations! Discretize.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.6084/m9.figshare.12751955.v1">doi:10.6084/m9.figshare.12751955.v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zhwjvt23tfhmjljetsfobsv5q4">fatcat:zhwjvt23tfhmjljetsfobsv5q4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200817222305/https://s3-eu-west-1.amazonaws.com/pfigshare-u-files/24131045/neural_differential_equations.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.6084/m9.figshare.12751955.v1"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> figshare.com </button> </a>

Graph Neural Ordinary Differential Equations [article]

Michael Poli, Stefano Massaroli, Junyoung Park, Atsushi Yamashita, Hajime Asama, Jinkyoo Park
<span title="2021-06-22">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
structures and differential equations.  ...  The proposed framework is shown to be compatible with various static and autoregressive GNN models.  ...  GDEs, on the other hand, closely track both positions and velocities of the particles.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.07532v4">arXiv:1911.07532v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4cb6fej2avacpbjhush7lhgxum">fatcat:4cb6fej2avacpbjhush7lhgxum</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210624064542/https://arxiv.org/pdf/1911.07532v4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/85/40/8540780e6b9422f7a1264edb70f39d3ff79bb8c1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.07532v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Deep Graph Matching via Blackbox Differentiation of Combinatorial Solvers [article]

Michal Rolínek, Paul Swoboda, Dominik Zietlow, Anselm Paulus, Vít Musil, Georg Martius
<span title="2020-08-05">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In addition, we highlight the conceptual advantages of incorporating solvers into deep learning architectures, such as the possibility of post-processing with a strong multi-graph matching solver or the  ...  Using the presence of heavily optimized combinatorial solvers together with some improvements in architecture design, we advance state-of-the-art on deep graph matching benchmarks for keypoint correspondence  ...  The optimizer in use is Adam [31] with an initial learning rate of 2 × 10 −3 which is halved four times in regular intervals. Learning rate for finetuning the VGG weights is multiplied with 10 −2 .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.11657v2">arXiv:2003.11657v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/7vmpyk23rjhjbkfioryiwisjc4">fatcat:7vmpyk23rjhjbkfioryiwisjc4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201012064612/https://arxiv.org/pdf/2003.11657v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/bd/c8/bdc825223d4199a714d3c357edbf0161174cafc0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.11657v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Learning Constrained Adaptive Differentiable Predictive Control Policies With Guarantees [article]

Jan Drgona, Aaron Tuor, Draguna Vrabie
<span title="2022-01-27">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We present differentiable predictive control (DPC), a method for learning constrained neural control policies for linear systems with probabilistic performance guarantees.  ...  We demonstrate that the proposed method can learn parametric constrained control policies to stabilize systems with unstable dynamics, track time-varying references, and satisfy nonlinear state and input  ...  Also, we want to thank our anonymous reviewers for their constructive feedback.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.11184v6">arXiv:2004.11184v6</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/sem2kvkrt5cnlhz2hgywkpmrge">fatcat:sem2kvkrt5cnlhz2hgywkpmrge</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220130223701/https://arxiv.org/pdf/2004.11184v6.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/07/ab/07ab147baac2b52e4acc2aad692ee97764c68dba.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.11184v6" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Adaptive-Control-Oriented Meta-Learning for Nonlinear Systems [article]

Spencer M. Richards, Navid Azizan, Jean-Jacques Slotine, Marco Pavone
<span title="2021-06-19">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our key insight is that we can better prepare the controller for deployment with control-oriented meta-learning of features in closed-loop simulation, rather than regression-oriented meta-learning of features  ...  Specifically, we meta-learn the adaptive controller with closed-loop tracking simulation as the base-learner and the average tracking error as the meta-objective.  ...  ACKNOWLEDGEMENTS We thank Masha Itkina for her invaluable feedback, and Matteo Zallio for his expertise in crafting Figure 1 .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2103.04490v2">arXiv:2103.04490v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/uf7fzau2qjc6fk2oco3tjp73ku">fatcat:uf7fzau2qjc6fk2oco3tjp73ku</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210623234044/https://arxiv.org/pdf/2103.04490v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/fd/65/fd659a30a40b9056b9965f0d130cf0e9cc0c2b23.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2103.04490v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

DiffEqFlux.jl - A Julia Library for Neural Differential Equations [article]

Chris Rackauckas, Mike Innes, Yingbo Ma, Jesse Bettencourt, Lyndon White, Vaibhav Dixit
<span title="2019-02-06">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We conclude by discussing the various adjoint methods used for backpropogation of the differential equation solvers.  ...  DiffEqFlux.jl is a library for fusing neural networks and differential equations.  ...  With access to the full range of solvers for ODEs, SDEs, DAEs, DDEs, PDEs, discrete stochastic equations, and more, we are interested to see what kinds of next generation neural networks you will build  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1902.02376v1">arXiv:1902.02376v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/gfd7spc3xngttiwpqun7nslkly">fatcat:gfd7spc3xngttiwpqun7nslkly</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200830022931/https://arxiv.org/pdf/1902.02376v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/70/a3/70a3900b6c74b1dfffb0920e3a49ab5dd1b80a59.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1902.02376v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Control-oriented meta-learning [article]

Spencer M. Richards, Navid Azizan, Jean-Jacques Slotine, Marco Pavone
<span title="2022-04-14">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our key insight is that we can better prepare the controller for deployment with control-oriented meta-learning of features in closed-loop simulation, rather than regression-oriented meta-learning of features  ...  when deployed in closed-loop for trajectory tracking control.  ...  Acknowledgements We thank Masha Itkina for her invaluable feedback, and Matteo Zallio for his expertise in crafting Figure 1 .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2204.06716v1">arXiv:2204.06716v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/pntvtsi7xbaynilchiun2za3tu">fatcat:pntvtsi7xbaynilchiun2za3tu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220507184537/https://arxiv.org/pdf/2204.06716v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/2f/51/2f51e4a1538e5cf0e4bdcc34c33a42efc22a0488.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2204.06716v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Impedance modelling on multiple dielectric builds

Martyn Gaudion, J. Alan Staniforth
<span title="">2004</span> <i title="Emerald"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/tficc7pftjhcpdqup3yhf4ocka" style="color: black;">Circuit world</a> </i> &nbsp;
Figure 16 shows a differential surface microstrip with a solder mask where allowance has been made for different mask thickness above the track, outside the tracks, and between the tracks.  ...  Figure 15 shows the boundary con guration used for a single track surface microstrip with two substrates below the track.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1108/03056120410512091">doi:10.1108/03056120410512091</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zlix2are45e55fqx5hnhv3qr5u">fatcat:zlix2are45e55fqx5hnhv3qr5u</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170811200134/http://www.polarinstruments.com/support/cits/Gaudion_CW_2004_30.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/3f/82/3f8286178c1021a08ecea0da39a99f93a320f9c7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1108/03056120410512091"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Learned discretizations for passive scalar advection in a 2-D turbulent flow [article]

Jiawei Zhuang, Dmitrii Kochkov, Yohai Bar-Sinai, Michael P. Brenner, Stephan Hoyer
<span title="2020-11-05">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The machine learning component is tightly integrated with traditional finite-volume schemes and can be trained via an end-to-end differentiable programming framework.  ...  Here we use a machine learning approach to learn a numerical discretization that retains high accuracy even when the solution is under-resolved with classical methods.  ...  In contrast, our neural network model closely tracks the reference "true solution" obtained by the 8× resolution baseline.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.05477v2">arXiv:2004.05477v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/isjqywlizrf4fl2434t2mcednq">fatcat:isjqywlizrf4fl2434t2mcednq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201110001900/https://arxiv.org/pdf/2004.05477v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/67/a4/67a4dd5210b8102b5bdbf0eb6e23b8ea2ecc71ce.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.05477v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Neural Non-Rigid Tracking [article]

Aljaž Božič, Pablo Palafox, Michael Zollhöfer, Angela Dai, Justus Thies, Matthias Nießner
<span title="2021-01-12">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
By enabling gradient back-propagation through the weighted non-linear least squares solver, we are able to learn correspondences and confidences in an end-to-end manner such that they are optimal for the  ...  We introduce a novel, end-to-end learnable, differentiable non-rigid tracker that enables state-of-the-art non-rigid reconstruction by a learned robust optimization.  ...  prediction network used for non-rigid tracking of two frames; • a self-supervised approach for learned correspondence weighting, which is informed by our differentiable solver and enables efficient, robust  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.13240v2">arXiv:2006.13240v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4jwandxiebav3ofgwdxkyto6ki">fatcat:4jwandxiebav3ofgwdxkyto6ki</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210115213855/https://arxiv.org/pdf/2006.13240v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c4/1d/c41dd2e239a6217869c2fec22dd6fbbc99cb992b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.13240v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Meta Learning with Differentiable Closed-form Solver for Fast Video Object Segmentation [article]

Yu Liu, Lingqiao Liu, Haokui Zhang, Hamid Rezatofighi, Ian Reid
<span title="2019-09-28">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our proposed meta-learning method uses a closed form optimizer, the so-called "ridge regression", which has been shown to be conducive for fast and better training convergence.  ...  In this paper, we tackle this task by formulating it as a meta-learning problem, where the base learner grasping the semantic scene understanding for a general type of objects, and the meta learner quickly  ...  Ridge Regression Ridge regression is a closed form solver and widely-used in machine learning community [34, 27] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.13046v1">arXiv:1909.13046v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/jygpxvlosbg7fl3w737x7iduai">fatcat:jygpxvlosbg7fl3w737x7iduai</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200929040005/https://arxiv.org/pdf/1909.13046v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/38/8a/388ab8ff4f2ccb53064297da438fecded2f61a26.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1909.13046v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Deep Time-Stream Framework for Click-Through Rate Prediction by Tracking Interest Evolution [article]

Shu-Ting Shi, Wenhao Zheng, Jun Tang, Qing-Guo Chen, Yao Hu, Jianke Zhu, Ming Li
<span title="2020-01-08">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recently, deep learning models have been proposed to learn the representation of users' overall interests, while ignoring the fact that interests may dynamically change over time.  ...  Experiments on public dataset as well as real industry dataset with billions of samples demonstrate the effectiveness of proposed approaches, which achieve superior performance compared with existing methods  ...  For Profile features, user profiles such as gender, age, activity score are used. Training dataset is generated with k = 1, 2, · · · , N − 2 for each user.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2001.03025v1">arXiv:2001.03025v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/e4sg36fbi5affixyd475bqd32u">fatcat:e4sg36fbi5affixyd475bqd32u</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200321041314/https://arxiv.org/pdf/2001.03025v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2001.03025v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Probabilistic Solutions to Differential Equations and their Application to Riemannian Statistics [article]

Philipp Hennig, Søren Hauberg
<span title="2014-02-12">2014</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our approach is an argument for a wider point that uncertainty caused by numerical calculations should be tracked throughout the pipeline of machine learning algorithms.  ...  Such methods have concrete value in the statistics on Riemannian manifolds, where non-analytic ordinary differential equations are involved in virtually all computations.  ...  Hauberg is supported in part by the Villum Foundation as well as an AWS in Education Machine Learning Research Grant award from Amazon.com.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1306.0308v2">arXiv:1306.0308v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6o2yo2dwovdbrb5wqd4jdoo3ma">fatcat:6o2yo2dwovdbrb5wqd4jdoo3ma</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200902121212/https://arxiv.org/pdf/1306.0308v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/b4/12/b412ff7257d8b904007843c4caa9ed889dbe463e.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1306.0308v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 11,797 results