A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf
.
Filters
Mirror Descent and the Information Ratio
[article]
2020
arXiv
pre-print
We establish a connection between the stability of mirror descent and the information ratio by Russo and Van Roy [2014]. ...
Our analysis shows that mirror descent with suitable loss estimators and exploratory distributions enjoys the same bound on the adversarial regret as the bounds on the Bayesian regret for information-directed ...
Connections between the information ratio and mirror descent were investigated by Zimmert and Lattimore [2019] , who showed that bounds on the stability of mirror descent imply bounds on the information ...
arXiv:2009.12228v1
fatcat:ai56kz53pvasfe3desuu655nja
First-Order Bayesian Regret Analysis of Thompson Sampling
[article]
2022
arXiv
pre-print
In this framework, Russo and Van Roy proposed an information-theoretic analysis of Thompson Sampling based on the information ratio, resulting in optimal worst-case regret bounds. ...
Finally, we introduce a novel link between Bayesian agents and frequentist confidence intervals. ...
See also [LS19] which improves the Russo and Van Roy entropic bound using Tsallis entropy, and [LG21] which further elucidates the connection between generalized information ratios and mirror descent ...
arXiv:1902.00681v3
fatcat:2gripwaj6nerbbd5tbyhua4qdy
An Information-Theoretic Analysis of Thompson Sampling
[article]
2015
arXiv
pre-print
This analysis inherits the simplicity and elegance of information theory and leads to regret bounds that scale with the entropy of the optimal-action distribution. ...
We provide an information-theoretic analysis of Thompson sampling that applies across a broad range of online optimization problems in which a decision-maker must learn from partial feedback. ...
Inspired by a connection between Thompson sampling and exponential weighting schemes, that paper introduced a family of Thompson sampling like algorithms and studied their application to contextual bandit ...
arXiv:1403.5341v2
fatcat:ilkglh4n2jacrbcobximevgme4
Discovering Boolean Gates in Slime Mould
[article]
2016
arXiv
pre-print
The search for gates was performed by both sweeping across configurations in the real material as well as training a neural network-based model and searching the gates therein using gradient descent. ...
We apply arbitrary electrical signals to 'configure' the slime mould, i.e. change shape of its body and, measure the slime mould's electrical response. ...
Simon and Andy prepared the mould and performed the exhaustive search experiments on the hardware platform, Jan, Klaus and Jürgen contributed with the mould neural network modelling. ...
arXiv:1607.02168v1
fatcat:cihaptlg5rcz7d7rn2wppjicpy
A lower bound on the detectability of nonassociative learning in the local bending reflex of the medicinal leech
1993
Behavioral and Neural Biology
The model comprised 4 sensory neurons, 10 to 40 interneurons, 8 motor neurons, and up to 480 connections. ...
Synaptic connections in the model were first optimized to reproduce the amplitude and time course of motor neuron synaptic potentials recorded during local bending. ...
proposed by Groves and Thompson (1970) . ...
doi:10.1016/0163-1047(93)90974-m
pmid:8503826
fatcat:tjgvsd77dfaupblg5bzqi7ar4a
Online convex optimization and no-regret learning: Algorithms, guarantees and applications
[article]
2018
arXiv
pre-print
Spurred by the enthusiasm surrounding the "Big Data" paradigm, the mathematical and algorithmic tools of online optimization have found widespread use in problems where the trade-off between data exploration ...
Particular attention is devoted to identifying the algorithms' theoretical performance guarantees and to establish links with classic optimization paradigms (both static and stochastic). ...
mirror descent. ...
arXiv:1804.04529v1
fatcat:2vqsvkhjmndyjeetb2tblxskia
D'Arcy W. Thompson's Cartesian transformations: a critical evaluation
2020
Zoomorphology
The images of D'Arcy Wentworth Thompson's book "On Growth and Form" got an iconic status and became influential for biometrics and other mathematical approaches to organismic form. ...
Based on his approach, Thompson formulated far-reaching conclusions with a partly anti-Darwinian stance. ...
We thank Stefan Zachow and Carola Becker for valuable discussions and Till Ramm for help with the scanning of crab carapaces. ...
doi:10.1007/s00435-020-00494-1
fatcat:hpok5vyuuzgsnkfifbipb3lsay
Zooglider: An autonomous vehicle for optical and acoustic sensing of zooplankton
2018
Limnology and Oceanography : Methods
The Zoocam quantifies zooplankton and marine snow as they flow through a defined volume inside a sampling tunnel. ...
The Zonar permits differentiation of large and small acoustic backscatterers in larger volumes than can be sampled optically. ...
Whitmore, by the Extreme Science and Engineering Discovery Environment (XSEDE) via National Science Foundation grant ACI-1548562, and indirectly by NSF via the California Current Ecosystem LTER site. ...
doi:10.1002/lom3.10301
fatcat:sywqmhdfezabndx4rqwrcumxe4
Gradual emergence of spontaneous correlated brain activity during fading of general anesthesia in rats: Evidences from fMRI and local field potentials
2015
NeuroImage
In the present study, we induced a deep state of anesthesia in rats by means of a ketamine/medetomidine peritoneal injection, and analyzed the time course of the correlation between the brain activity ...
sparse connectivity, stereotypical activity and poor functional integration, during lighter states different frequency-specific functional networks emerge, endowing the gradual restoration of structured ...
25 min each, i.e. long enough to sample substantial and reliable differences between brain states. ...
doi:10.1016/j.neuroimage.2015.03.037
pmid:25804643
pmcid:PMC4461308
fatcat:ix63hls3pfax3oerrk76aocdma
Reinforcement Learning, Bit by Bit
[article]
2022
arXiv
pre-print
The design of data-efficient agents calls for a deeper understanding of information acquisition and representation. We discuss concepts and regret analysis that together offer principled guidance. ...
This line of thinking sheds light on questions of what information to seek, how to seek that information, and what information to retain. ...
Acknowledgments Our thinking about the relation between information and sequential decision was shaped by an earlier collaboration with Dan Russo, which focused on bandit environments. ...
arXiv:2103.04047v7
fatcat:gpc3cvdk2nekpg42ny6n2w7i7m
Distilled Thompson Sampling: Practical and Efficient Thompson Sampling via Imitation Learning
[article]
2020
arXiv
pre-print
Thompson sampling (TS) has emerged as a robust technique for contextual bandit problems. ...
The explicit policy representation enables fast online decision-making and easy deployment in mobile and server-based environments. ...
The contextual information accompanying each video includes dense and sparse features about: the video file (e.g. the raw bitrate, resolution, and file size) and the network connection (e.g. connection ...
arXiv:2011.14266v2
fatcat:fspq3k7trffy5lan2p4pevlzyq
Genetic analysis reveals population structure and recent migration within the highly fragmented range of the Cross River gorilla (Gorilla gorilla diehli)
2006
Molecular Ecology
and one-third of the estimated total population through the use of noninvasively collected DNA samples. ...
Conservation efforts should strive to maintain connectivity between subpopulations that are still in migratory contact and attempt to restore connectivity where it has been lost. ...
Oates for help developing and implementing the project and providing helpful comments on the manuscript. We also thank J. Sunderland ...
doi:10.1111/j.1365-294x.2006.03159.x
pmid:17257109
fatcat:22t2gkh6nfc4xm55c3vvgf5swy
Spatial fine-mapping for gene-by-environment effects identifies risk hot spots for schizophrenia
2018
Nature Communications
We analyzed a case-cohort study (n = 24,028), drawn from the 1.47 million Danish persons born between 1981 and 2005, using a novel framework for decomposing the geospatial risk for schizophrenia based ...
Effects of specific locales vary substantially within the most densely populated city of Denmark, with hazard ratios ranging from 0.26 to 9.26 for environment and from 0.20 to 5.95 for gene-by-environment ...
(Big Data Center for Environment and Health, grant number NNF17OC0027864), and conducted using the Danish National Biobank resource supported by the Novo Nordisk Foundation. ...
doi:10.1038/s41467-018-07708-7
pmid:30546018
pmcid:PMC6294340
fatcat:k6kyo2oxhvbidorveyl3a5gw54
Walking infants adapt locomotion to changing body dimensions
2000
Journal of Experimental Psychology: Human Perception and Performance
Overall, infants' exploratory activity mirrored their go ratios (see Figure 7) , suggesting that they used information from looking, swaying, and touching movements to judge whether slopes were safe for ...
and haptic information from torque at the ankles and from shearing forces between the bottom of the foot and the sloping surface). ...
Substituting infants' mass and the location of their center of mass in each condition into Equation 3, the ratio of sin0~,JsinO/~,uh = .75 when infants swayed around their ankles, thus reducing the angle ...
doi:10.1037//0096-1523.26.3.1148
pmid:10884014
fatcat:ff4mei67knfpbovzejxzrqgigy
Cassini/Huygens Science Instruments, Spacecraft, and Mission
1997
Journal of Spacecraft and Rockets
S., and Thompson, J. ...
S-, X-, and Ka-band link mea- surements between orbiter and Earth provide information about in- tervening material and gravity fields. ...
doi:10.2514/2.3241
fatcat:ibkpa7jjzvav7l53zgt2f4ekvm
« Previous
Showing results 1 — 15 out of 1,647 results