Filters








10,835 Hits in 2.7 sec

Learning rotations with little regret

Elad Hazan, Satyen Kale, Manfred K. Warmuth
2016 Machine Learning  
We also show that any deterministic algorithm for learning rotations has Ω(T ) regret in the worst case.  ...  We describe online algorithms for learning a rotation from pairs of unit vectors in R n .  ...  Acknowledgement Our work was motivated by preliminary research done with Adam Smith [SW10, SW08] and we greatly benefited from discussions with him.  ... 
doi:10.1007/s10994-016-5548-x fatcat:v7azubwm6zcbfkcccbu7jhntby

Enforcing Predictive Invariance across Structured Biomedical Domains [article]

Wengong Jin, Regina Barzilay, Tommi Jaakkola
2020 arXiv   pre-print
oracle with hindsight access to held-out environments.  ...  To address these challenges, we propose a new regret minimization (RGM) algorithm and its extension for structured environments.  ...  The proposed approach enables us to effectively utilize readily available, heterogeneous data to model bioactivity, reducing prohibitive cost and time associated with traditional drug discovery workflow  ... 
arXiv:2006.03908v3 fatcat:spxtfllivvam3gkuqtq5pjdoq4

A little knowledge . .

2016 In practice (London. 1979)  
This is certainly the case at vet schoolcoming off rotations and suddenly noticing conditions that you had been previously blind to.  ...  A Practitioner Ponders AS a recent graduate, a little knowledge is, despite the best efforts of my vet school's fi ne clinicians, at best what I have.  ... 
doi:10.1136/inp.i5178 fatcat:yr4adwi4cnfm3mlpajeudest2q

The impact of doubt on the experience of regret

Philippe P.F.M. van de Calseyde, Marcel Zeelenberg, Ellen R.K. Evers
2018 Organizational Behavior and Human Decision Processes  
Varimax rotation.  ...  At stage 1, more regret before knowing the outcome (i.e., preoutcome regret) was significantly associated with more relief after learning the outcome.  ...  Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.obhdp.2018.08.006.  ... 
doi:10.1016/j.obhdp.2018.08.006 fatcat:33xp2dunevaffpnyeakeyuc2wm

Modern X-ray Analysis on Single Crystals. A Practical Guide. Second edition. By Peter Luger. De Gruyter, 2014. Pp. xi+334. Price EUR 119.95/USD 168.00/GBP 89.99. ISBN 978-3-11-030823-5

Claude Lecomte
2016 Acta Crystallographica Section A: Foundations and Advances  
The theory on four-circle single-crystal diffractometers is mostly limited to point detector data collection with too little on area detectors, in particular: how to estimate the standard deviations on  ...  The first edition of this book was widely recognized as one of the best X-ray crystallography books for anyone starting crystallography to learn organic/organometallic single-crystal structure determination  ...  The theory on four-circle single-crystal diffractometers is mostly limited to point detector data collection with too little on area detectors, in particular: how to estimate the standard deviations on  ... 
doi:10.1107/s205327331502121x fatcat:jdlcaa4eibbfvau4bypubm2454

LXXXIII. On the passage of the comet of 1819 across the disc of the sun

M. Olbers
1821 The Philosophical Magazine  
But on approaching the same part of the wire, with the south pole of the magnet, the apparatus will rotate by repulsion. 5.  ...  Prolbssor Brandes at Breslau viewed the sun on the 26th of June, a little before midday, with a glass of a thirty-four times multiplying power, and he perceived a ~pot distinctly visible, almost passing  ... 
doi:10.1080/14786442108652545 fatcat:lqquijl6hncabasfvdqbj7fhke

Graph Neural Network Guided Local Search for the Traveling Salesperson Problem [article]

Benjamin Hudson and Qingbiao Li and Matthew Malencia and Amanda Prorok
2021 arXiv   pre-print
Our model predicts the regret of including each edge of the problem graph in the solution; GLS uses these predictions in conjunction with the original problem graph to find solutions.  ...  When generalizing from 20-node problems to the 100-node problem set, our approach finds solutions with an average optimality gap of 2.5%, a 10x improvement over the next best learning-based benchmark.  ...  We train the model for 100 epochs with early stopping.  ... 
arXiv:2110.05291v2 fatcat:wncbcol76nbqhaflcw2ewhafvm

Learning Active Task-Oriented Exploration Policies for Bridging the Sim-to-Real Gap [article]

Jacky Liang and Saumya Saxena and Oliver Kroemer
2020 arXiv   pre-print
We instantiate the framework in simulation with the Linear Quadratic Regulator as well as in the real world with pouring and object dragging tasks.  ...  The latter requires learning policies that concurrently perform the task and generate useful trajectories for system identification.  ...  The method in [3] combines this idea with Passive Sys-Id to produce a model that can iteratively reduce the residual errors with little real-world data.  ... 
arXiv:2006.01952v2 fatcat:waymaogxjzarrhunsgtz567fjm

Behavioral Specification of the Entrustment Process

Anna T. Cianciolo, Jason A. Kegg
2013 Journal of Graduate Medical Education  
The loner takes complete control of the situation, with the entrustment decision offering the trainee little to no independence, even in relatively safe settings.  ...  Yet, it is the essence of clinical teaching and practicebased learning and improvement.  ...  The loner takes complete control of the situation, with the entrustment decision offering the trainee little to no independence, even in relatively safe settings.  ... 
doi:10.4300/jgme-d-12-00158.1 pmid:24404219 pmcid:PMC3613292 fatcat:cbgymygur5bstif5af3hgrrouy

Adaptive Metamorphic Testing with Contextual Bandits [article]

Helge Spieker, Arnaud Gotlieb
2020 arXiv   pre-print
We present experimental results over two major case studies in machine learning, namely image classification and object detection, and identify weaknesses and robustness boundaries.  ...  By using contextual bandits, Adaptive Metamorphic Testing learns which metamorphic relations are likely to transform a source test case, such that it has higher chance to discover faults.  ...  The exploration algorithm then chooses from those actions which have not been learned to perform bad, i.e., having high regret, in the current context.  ... 
arXiv:1910.00262v3 fatcat:6y25b53dlffqxovcsyret7snqe

Page 341 of Ballou's Monthly Magazine Vol. 69, Issue 4 [page]

1889 Ballou's Monthly Magazine  
But the tale is perfectly true, and both the gentlemen who saw it are still alive; and I only regret that it was not my good luck to see the affair with my own eyes.  ...  The velocity of its rotation is known.  ... 

Robust Tracking with Weighted Online Structured Learning [chapter]

Rui Yao, Qinfeng Shi, Chunhua Shen, Yanning Zhang, Anton van den Hengel
2012 Lecture Notes in Computer Science  
We thus also propose a principled approach to weighted online learning using weighted reservoir sampling and provide a weighted regret bound as a theoretical guarantee of performance.  ...  One of the difficulties with the online learning approach to this problem has been a lack of flexibility in the modelling of the inevitable variations in target and scene appearance over time.  ...  [20] introduced a regret for online learning for binary classification. Bordes et al. [19] extended the regret bound to structured online learning.  ... 
doi:10.1007/978-3-642-33712-3_12 fatcat:uclm2gio45hn5jqmhgspolm4eq

Not-So-Random Features [article]

Brian Bullins, Cyril Zhang, Yi Zhang
2018 arXiv   pre-print
We propose a principled method for kernel learning, which relies on a Fourier-analytic characterization of translation-invariant or rotation-invariant kernels.  ...  In our experiments, we find that the step size achieving the standard regret guarantee (scaling as 1/ √ T ) tends to be a little too conservative.  ...  Intriguingly, our analysis interprets the main algorithm as no-regret learning dynamics in a zero-sum minmax game, whose value is the classification margin.  ... 
arXiv:1710.10230v2 fatcat:2hxaw7bverbrdivp4uujih5dqq

Recent Literature A Dictionary of Medicine . Including General Pathology, General Therapeutics, Hygiene and the Diseases of Women and Children. By various writers. Edited by Richard Quain, Bart., M.D., LL.D., F.R.S., etc., assisted by Frederick Thomas Roberts, M.D., B.Sc., etc., and J. Mitchell Bruce, M.A., M.D., etc. With an American Appendix by Samuel Treat Armstrong, M.D., Ph.D., etc. New edition, revised throughout and enlarged. Vol. I, Abdomen — Lysis; Vol. II, Macrocheilia —Zyme. New York: D. Appleton ...

1894 Boston Medical and Surgical Journal  
The chapter on pelvimetry is timely and excellent, though we disagree with the statement that the justo-minor pelvis is a rare  ...  It is to be regretted that the section on the use of the forceps falls a little below the very high standard of the rest of the book.  ...  We dissent heartily from tbe author's statement that most lacerations of the cervix should be immediately repaired ; though we agree with him as to the ease with which the operation can be performed with  ... 
doi:10.1056/nejm189412131312410 fatcat:mmauz3246vcqxj4brnmcxlsnky

Recent Literature A Treatise on the Principles and Practice of Medicine . Designed for the use of Practitioners and Students of Medicine. By Austin Flint, M.D., LL.D. Seventh edition, thoroughly revised. By Frederick P. Henry, A.M., M.D. Philadelphia: Lea Brothers & Co. 1894

1894 Boston Medical and Surgical Journal  
The chapter on pelvimetry is timely and excellent, though we disagree with the statement that the justo-minor pelvis is a rare  ...  It is to be regretted that the section on the use of the forceps falls a little below the very high standard of the rest of the book.  ...  We dissent heartily from tbe author's statement that most lacerations of the cervix should be immediately repaired ; though we agree with him as to the ease with which the operation can be performed with  ... 
doi:10.1056/nejm189412131312409 fatcat:sqkbejalpraavixpc2iwcjbn7m
« Previous Showing results 1 — 15 out of 10,835 results