scholarly journals The Perils of Misspecified Priors and Optional Stopping in Multi-Armed Bandits

2021 ◽  
Vol 4 ◽  
Author(s):  
Markus Loecher
Keyword(s):  
Posterior Probability ◽  
Prior Probability ◽  
Strong Dependence ◽  
Stopping Times ◽  
Bayesian Posterior Probability ◽  
Thompson Sampling ◽  
Optional Stopping ◽  
The Subject ◽  
Active Research ◽  
Optimal Stopping Times

The connection between optimal stopping times of American Options and multi-armed bandits is the subject of active research. This article investigates the effects of optional stopping in a particular class of multi-armed bandit experiments, which randomly allocates observations to arms proportional to the Bayesian posterior probability that each arm is optimal (Thompson sampling). The interplay between optional stopping and prior mismatch is examined. We propose a novel partitioning of regret into peri/post testing. We further show a strong dependence of the parameters of interest on the assumed prior probability density.

scholarly journals Are Multi-Armed Bandits Susceptible to Peeking?

2018 ◽  
Vol 21 (1) ◽  
pp. 95-104
Author(s):  
Markus Loecher
Keyword(s):  
Standard Method ◽  
Bayesian Methods ◽  
Web Sites ◽  
Multiple Testing ◽  
Posterior Probability ◽  
Prior Probability ◽  
Strong Dependence ◽  
P Value ◽  
Repeated Testing ◽  
Parameters Of Interest

Abstract A standard method to evaluate new features and changes to e.g. Web sites is A/B testing. A common pitfall in performing A/B testing is the habit of looking at a test while it’s running, then stopping early. Due to the implicit multiple testing, the p-value is no longer trustworthy and usually too small. We investigate the claim that Bayesian methods, unlike frequentist tests, are immune to this “peeking” problem. We demonstrate that two regularly used measures, namely posterior probability and value remaining are severely affected by repeated testing. We further show a strong dependence on the prior probability of the parameters of interest.

scholarly journals An Impact Location Algorithm for Spacecraft Stiffened Structure Based on Posterior Possibility Correlation

Sensors ◽  
2020 ◽  
Vol 20 (2) ◽  
pp. 368 ◽  
Author(s):  
Lei Qi ◽  
Zhoumo Zeng ◽  
Lichen Sun ◽  
Xiaobo Rui ◽  
Fan Fan ◽  
...  
Keyword(s):  
Posterior Probability ◽  
Cross Correlation ◽  
Bayesian Posterior Probability ◽  
Probability Method ◽  
Stiffened Plate ◽  
Plate Structures ◽  
Impact Location ◽  
Weight Method ◽  
Location Algorithm ◽  
The Impact

In order to ensure the safety of spacecrafts in orbit, impact location is an important part of structural health monitoring systems. In this paper, an impact location algorithm based on posterior probability correlation is proposed to solve the problem, that is, the impact point in the stiffened structure of a spacecraft is difficult to locate. The algorithm combines the Gaussian cross-correlation possibility weight method and the Bayesian posterior probability method. The cross-correlation possibility weight superposition based on grids was used to reduce the dependence of the accuracy of time difference extraction. Gaussian and normalized fitting were used to compensate the reflection, modal transformation, and amplitude attenuation of a stiffened plate. The location result was further optimized by the posterior probability. The proposed algorithm can be applied to the impact source localization of complex stiffened plate structures. The experiment results showed that the average location error can be 2.57 cm with proper sensor network schemes.

Interfaces In Glass-Containing Ceramics

1999 ◽  
Vol 5 (S2) ◽  
pp. 98-99
Author(s):  
C. Barry Carter ◽  
N. Ravishankar ◽  
Carsten Korte ◽  
M.P. Mallamaci
Keyword(s):  
Grain Boundaries ◽  
Strong Dependence ◽  
Solid Substrate ◽  
Special Boundaries ◽  
Intergranular Films ◽  
The Subject ◽  
Microscopy Techniques ◽  
Sintered Oxide ◽  
Microstructural Examination

The interaction of a liquid with a solid substrate has been the subject of intense research since the early 1800s. The subject is relevant and very important even today from a basic science as well as technological standpoint. Liquid-phase-sintered oxide ceramics often contain a siliceous glassy layer in the grain boundaries which affect the properties of the ceramic. The wetting of the grain boundaries by the liquid has a strong dependence on the crystallography. While some boundaries are preferably wet, other special boundaries appear to be completely ‘dry’. The presence of the liquid film also affects the faceting behavior of the boundary plane. While several studies have been carried out on these intergranular glass films, many of the basic issues still remain unclear. The development of modern microscopy techniques facilitate the characterization of structure and chemistry at a high spatial resolution. This paper deals with the microstructural examination of two of the important issues involved with these intergranular films, i.e., faceting of grain boundaries in the presence of a glassy phase and the behavior of interfaces between two glassy phases of different compositions.

Optimal stopping with discount and observation costs

2000 ◽  
Vol 37 (01) ◽  
pp. 64-72 ◽  
Author(s):  
Robert Kühne ◽  
Ludger Rüschendorf
Keyword(s):  
Differential Equations ◽  
Optimal Stopping ◽  
Point Processes ◽  
Numerical Solutions ◽  
Domain Of Attraction ◽  
Approximate Solutions ◽  
Poisson Approximation ◽  
Stopping Times ◽  
Optimal Stopping Problem ◽  
Optimal Stopping Times

For i.i.d. random variables in the domain of attraction of a max-stable distribution with discount and observation costs we determine asymptotic approximations of the optimal stopping values and asymptotically optimal stopping times. The results are based on Poisson approximation of related embedded planar point processes. The optimal stopping problem for the limiting Poisson point processes can be reduced to differential equations for the boundaries. In several cases we obtain numerical solutions of the differential equations. In some cases the analysis allows us to obtain explicit optimal stopping values. This approach thus leads to approximate solutions of the optimal stopping problem of discrete time sequences.

Paurodontella gilanica n. sp. (Nematoda: Sphaerularioidea) from Iran

Nematology ◽  
2018 ◽  
Vol 20 (5) ◽  
pp. 471-482 ◽  
Author(s):  
Ali Yaghoubi ◽  
Ebrahim Pourjam ◽  
Majid Pedram
Keyword(s):  
New Species ◽  
Posterior Probability ◽  
Small Body ◽  
Bootstrap Support ◽  
Bayesian Posterior Probability ◽  
Female Reproductive System ◽  
Lateral Field ◽  
Morphological Differences ◽  
Tail Tip ◽  
18S Rdna Sequence

Summary Paurodontella gilanica n. sp. is described and illustrated based on morphological, morphometric and molecular characters. The new species is characterised by its small body size, four lines in the lateral field, weak stylet with minute asymmetrical knobs, female reproductive system lacking a diverticulum but with small post-vulval uterine sac, elongate conoid tail with pointed, sometimes filiform, tip, males common with small spicules and cloacal bursa not reaching tail tip. Morphological differences between the new species and seven known species of the genus, namely P. asymmetrica, P. auriculata, P. balochistanica, P. minuta, P. myceliophaga, P. parapitica and P. persica, are discussed. The new species is also compared with four known species of Paurodontus having four lines in the lateral field. Using a 1123 nt long partial 18S rDNA sequence of the new species to reconstruct the phylogenetic relationships revealed that it formed a clade with members of the Sphaerulariidae and Paurodontidae. Using a 746 nt long partial sequence of the 28S rDNA D2-D3 segment revealed that P. gilanica n. sp. formed a clade with Abursanema iranicum in both Bayesian inference (BI) and maximum likelihood (ML) analyses with 0.99 Bayesian posterior probability (BPP) and 89% bootstrap support value (BS). The morphological affinities of Paurodontella and Paurodontus are discussed.

Optimal stopping with discount and observation costs

2000 ◽  
Vol 37 (1) ◽  
pp. 64-72 ◽  
Author(s):  
Robert Kühne ◽  
Ludger Rüschendorf
Keyword(s):  
Differential Equations ◽  
Optimal Stopping ◽  
Point Processes ◽  
Numerical Solutions ◽  
Domain Of Attraction ◽  
Approximate Solutions ◽  
Poisson Approximation ◽  
Stopping Times ◽  
Optimal Stopping Problem ◽  
Optimal Stopping Times

For i.i.d. random variables in the domain of attraction of a max-stable distribution with discount and observation costs we determine asymptotic approximations of the optimal stopping values and asymptotically optimal stopping times. The results are based on Poisson approximation of related embedded planar point processes. The optimal stopping problem for the limiting Poisson point processes can be reduced to differential equations for the boundaries. In several cases we obtain numerical solutions of the differential equations. In some cases the analysis allows us to obtain explicit optimal stopping values. This approach thus leads to approximate solutions of the optimal stopping problem of discrete time sequences.

scholarly journals Double Optimal Stopping in the Fishing Problem

2009 ◽  
Vol 46 (02) ◽  
pp. 415-428
Author(s):  
Anna Karpowicz
Keyword(s):  
Dynamic Programming ◽  
Cost Function ◽  
Utility Function ◽  
Optimal Stopping ◽  
Renewal Process ◽  
Fixed Time ◽  
Time Dependent ◽  
Stopping Times ◽  
The Difference ◽  
Optimal Stopping Times

In this paper we consider the following problem. An angler buys a fishing ticket that allows him/her to fish for a fixed time. There are two locations to fish at the lake. The fish are caught according to a renewal process, which is different for each fishing location. The angler's success is defined as the difference between the utility function, which is dependent on the size of the fish caught, and the time-dependent cost function. These functions are different for each fishing location. The goal of the angler is to find two optimal stopping times that maximize his/her success: when to change fishing location and when to stop fishing. Dynamic programming methods are used to find these two optimal stopping times and to specify the expected success of the angler at these times.

Sign in / Sign up

Export Citation Format

Share Document