Filters








5,287 Hits in 4.6 sec

Learning to Optimize: A Primer and A Benchmark [article]

Tianlong Chen, Xiaohan Chen, Wuyang Chen, Howard Heaton, Jialin Liu, Zhangyang Wang, Wotao Yin
2021 arXiv   pre-print
This article is poised to be the first comprehensive survey and benchmark of L2O for continuous optimization.  ...  Learning to optimize (L2O) is an emerging approach that leverages machine learning to develop optimization methods, aiming at reducing the laborious iterations of hand engineering.  ...  The soft-thresholding function takes the formula as η θ (z) = sign(z) · max(0, |z| − θ) Learning to Optimize: A Primer and A Benchmark .  ... 
arXiv:2103.12828v2 fatcat:c75y3wz6cngirb2zpugjk63ymq

Learning to Scaffold: Optimizing Model Explanations for Teaching [article]

Patrick Fernandes, Marcos Treviso, Danish Pruthi, André F. T. Martins, Graham Neubig
2022 arXiv   pre-print
Modern machine learning models are opaque, and as a result there is a burgeoning academic subfield on methods that explain these models' behavior.  ...  learn to simulate the original model.  ...  We are grateful to Nuno Sabino, Thales Bertaglia, Henrico Brum, and Antonio Farinhas for the participation in human evaluation experiments.  ... 
arXiv:2204.10810v1 fatcat:setal7zuizerdosagoim7hwml4

Learning Combinatorial Optimization on Graphs: A Survey with Applications to Networking [article]

Natalia Vesselinova, Rebecca Steinert, Daniel F. Perez-Ramirez, Magnus Boman
2020 arXiv   pre-print
We organize and compare the structures involved with learning to solve combinatorial optimization problems, with a special eye on the telecommunications domain and its continuous development of live and  ...  Existing approaches to solving combinatorial optimization problems on graphs suffer from the need to engineer each problem algorithmically, with practical problems recurring in many instances.  ...  [23] , and Murphy [24] provide an indepth study of machine learning. A brief primer to the area is provided by Simone et al.  ... 
arXiv:2005.11081v1 fatcat:ajqghcevqvdrvdlcrknxlzlqdi

Learning Combinatorial Optimization on Graphs: A Survey with Applications to Networking

Natalia Vesselinova, Rebecca Steinert, Daniel F. Perez-Ramirez, Magnus Boman
2020 IEEE Access  
We organize and compare the structures involved with learning to solve combinatorial optimization problems, with a special eye on the telecommunications domain and its continuous development of live and  ...  Existing approaches to solving combinatorial optimization problems on graphs suffer from the need to engineer each problem algorithmically, with practical problems recurring in many instances.  ...  Junaid Shuja, who coordinated the review process, and to the anonymous reviewers for their expeditious and helpful review comments received in preparation of the final version of the manuscript.  ... 
doi:10.1109/access.2020.3004964 fatcat:v7i7x6p77zfi7dntipxoiolily

Using Support Vector Machines to Learn How to Compile a Method

Ricardo Nabinger Sanchez, Jose Nelson Amaral, Duane Szafron, Marius Pirvu, Mark Stoodley
2010 2010 22nd International Symposium on Computer Architecture and High Performance Computing  
The solution proposed is to use a Support Vector Machine (SVM) to learn a model based on method features and on the measured compilation and execution times of the methods.  ...  An extensive exploration phase collects a set of example compilations to be used by the SVM to train the model. This paper reports on a work in progress.  ...  A Primer on Support Vector Machines This section presents a brief primer of SVMs, which are thoroughly described in the literature, to help the reader understand the remainder of the paper.  ... 
doi:10.1109/sbac-pad.2010.35 dblp:conf/sbac-pad/SanchezASPS10 fatcat:qaqh3ioy6vhe3ep37ymc724kme

Reinforcement Learning to Solve NP-hard Problems: an Application to the CVRP [article]

Leo Ardon
2022 arXiv   pre-print
In this paper, we evaluate the use of Reinforcement Learning (RL) to solve a classic combinatorial optimization problem: the Capacitated Vehicle Routing Problem (CVRP).  ...  Moreover, instead of trying to solve a specific instance of the problem, the RL algorithm learns the skills required to solve the problem.  ...  Acknowledgement I would like to thank my supervisor, Dr. Wolfram Wiesemann, whose help and guidance was invaluable to help me drive this research project in the right direction.  ... 
arXiv:2201.05393v1 fatcat:gmiu7aeieneihkvj5oubup4su4

Practical Transfer Learning for Bayesian Optimization [article]

Matthias Feurer, Benjamin Letham, Frank Hutter, Eytan Bakshy
2022 arXiv   pre-print
We develop a new hyperparameter-free ensemble model for Bayesian optimization that is a generalization of two existing transfer learning extensions to Bayesian optimization and establish a worst-case bound  ...  When hyperparameter optimization of a machine learning algorithm is repeated for multiple datasets it is possible to transfer knowledge to an optimization run on a new dataset.  ...  Acknowledgements Thanks to Till Varoquaux for support in developing the method, Sam Daulton for help with the implementation, Martin Wistuba for help with the TST-R and TAF baselines, and Rodolphe Jenatton  ... 
arXiv:1802.02219v3 fatcat:xydfsavivndjpibrhod5mz7xf4

Optimizing the molecular diagnosis of Covid-19 by combining RT-PCR and a pseudo-convolutional machine learning approach to characterize virus DNA sequences [article]

Juliana Carneiro Gomes, Leandro Honorato de S. Silva, Janderson Ferreira, Agostinho A. F. Junior, Allana Lais dos Santos Rocha, Leticia Castro, Nathalia R. C. da Silva, Bruno Jose T. Fernandes, Wellington Pinheiro dos Santos
2020 bioRxiv   pre-print
Therefore, the molecular diagnosis of Covid-19 can be optimized by combining RT-PCR and our pseudo-convolutional method to identify SARS-Cov-2 DNA sequences faster with higher specificity and sensitivity  ...  Through the proposed method, it is possible to identify virus sequences from a large database: 347,363 virus DNA sequences from 24 virus families and SARSCov-2.  ...  Our main objective is to optimize the RT-PCR, the benchmark for Covid-19 diagnosis.  ... 
doi:10.1101/2020.06.02.129775 fatcat:im77b2kjxrdfhdv5v62qqethtu

Marginally calibrated response distributions for end-to-end learning in autonomous driving [article]

Clara Hoffmann, Nadja Klein
2021 arXiv   pre-print
These learners must provide reliable uncertainty estimates for their predictions in order to meet safety requirements and initiate a switch to manual control in areas of high uncertainty.  ...  To ensure the scalability to large n regimes, we develop efficient estimation based on variational inference as a fast alternative to computationally intensive, exact inference via Hamiltonian Monte Carlo  ...  We train two end-to-end learners for highway driving on the comma2k19 data and quantify the uncertainty of the steering angle predictions using the IC-NLM and a number of state-of-the-art benchmark methods  ... 
arXiv:2110.01050v1 fatcat:iolauf4mirbyxhypclb4tdadjy

A Regret Minimization Approach to Iterative Learning Control [article]

Naman Agarwal, Elad Hazan, Anirudha Majumdar, Karan Singh
2021 arXiv   pre-print
Based on recent advances in non-stochastic control, we design a new iterative algorithm for minimizing planning regret that is more robust to model mismatch and uncertainty.  ...  We provide theoretical and empirical evidence that the proposed algorithm outperforms existing methods on several benchmarks.  ...  A primer on Pontryagin's principle in optimal control. Collegiate publishers, 2015.  ... 
arXiv:2102.13478v1 fatcat:a6wtiktq7nf5takiyzpmtzozra

Rule-based specification mining leveraging learning to rank

Zherui Cao, Yuan Tian, Tien-Duy B. Le, David Lo
2018 Automated Software Engineering : An International Journal  
In this work, we propose a learning to rank based approach that automatically learns a good combination of 38 interestingness measures.  ...  To deal with the problem of lack of and outdated specifications, rule-based specification mining approaches have been proposed.  ...  At each iteration, a weak ranker is learned to optimize the ordering of rule pairs in the training data given their weights.  ... 
doi:10.1007/s10515-018-0231-z fatcat:m5hwvhcbq5azddmnwbnmpho3a4

Stochastic Optimization for Large-scale Optimal Transport [article]

Genevay Aude, Marco Cuturi, Gabriel Peyré
2016 arXiv   pre-print
This is currently the only known method to solve this problem, apart from computing OT on finite samples. We backup these claims on a set of discrete, semi-discrete and continuous benchmark problems.  ...  We propose a new class of stochastic optimization algorithms to cope with large-scale problems routinely encountered in machine learning applications.  ...  The work of A. Genevay has been supported by Région Ile-de-France. M. Cuturi gratefully acknowledges the support of JSPS young research A grant 26700002.  ... 
arXiv:1605.08527v1 fatcat:ip4qpnkvurgbxn6g37426z4qd4

A deep learning approach to pattern recognition for short DNA sequences [article]

Akosua Busia, George E. Dahl, Clara Fannjiang, David H. Alexander, Elizabeth Dorfman, Ryan Poplin, Cory Y. McLean, Pi-Chuan Chang, Mark DePristo
2018 bioRxiv   pre-print
Our results are a first step towards our long-term goal of developing a general-purpose deep learning model that can learn to predict any type of label from short biological sequences.  ...  In this work we describe a deep learning approach to solve such problems in a single step by training a deep neural network (DNN) to predict the database-derived labels directly from the query sequence  ...  Acknowledgements For their early input, which helped to frame the initial problem and understand potential applications, we thank Adam Roberts, Cinjon Resnick, and C. Rob Young.  ... 
doi:10.1101/353474 fatcat:4tivyjx74rhenkghp7efvt7wbe

The Limits to Learning a Diffusion Model [article]

Jackie Baek, Vivek F. Farias, Andreea Georgescu, Retsef Levi, Tianyi Peng, Deeksha Sinha, Joshua Wilde, Andrew Zheng
2021 arXiv   pre-print
We show that one cannot hope to learn such models until quite late in the diffusion.  ...  Specifically, we show that the time required to collect a number of observations that exceeds our sample complexity lower bounds is large.  ...  Acknowledgements The authors express their gratitude to Danial Mirza, Suzana Iacob, El Ghali Zerhouni, Neil Sanjay Pendse, Shen Chen, Celia Escribe and Jonathan Amar for their excellent support in data  ... 
arXiv:2006.06373v2 fatcat:y5usbbdenvdonc7v3khru5jnia

On Engagement: Learning to Pay Attention

R. Lisle Baker, Daniel P. Brown
2013 Social Science Research Network  
In an age of electronic and mental distraction, the ability to pay attention is a fundamental legal skill increasingly important for law students and the lawyers and judges they will become, not only for  ...  time and energy otherwise lost to internal or external distraction.  ...  Robert Altobello, Concentration and Contemplation: A Lesson in Learning to Learn, 5 J. TRANSFORMATIVE EDUC. 354, 368 (2007). 20. See Diane M.  ... 
doi:10.2139/ssrn.2269726 fatcat:3lplbppsgzhr7gzluqmawgk2o4
« Previous Showing results 1 — 15 out of 5,287 results