Search for: All records

Change‐point detection studies the problem of detecting the changes in the underlying distribution of the data stream as soon as possible after the change happens. Modern large‐scale, high‐dimensional, and complex streaming data call for computationally (memory) efficient sequential change‐point detection algorithms that are also statistically powerful. This gives rise to a computation versus statistical power trade‐off, an aspect less emphasized in the past in classic literature. This tutorial takes this new perspective and reviews several sequential change‐point detection procedures, ranging from classic sequential change‐point detection algorithms to more recent non‐parametric procedures that consider computation, memory efficiency, and model robustness in the algorithm design. Our survey also contains classic performance analysis, which provides useful techniques for analyzing new procedures.

This article is categorized under:

Statistical Models > Time Series Models

Algorithms and Computational Methods > Algorithms

Data: Types and Structure > Time Series, Stochastic Processes, and Functional Data

 

We present the first provably convergent two timescale off-policy actor-critic algorithm (COFPAC) with function approximation. Key to COFPAC is the introduction of a new critic, the emphasis critic, which is trained via Gradient Emphasis Learning (GEM), a novel combination of the key ideas of Gradient Temporal Difference Learning and Emphatic Temporal Difference Learning. With the help of the emphasis critic and the canonical value function critic, we show convergence for COF-PAC, where the critics are linear, and the actor can be nonlinear. 

The ratio of the electric to magnetic form factors of the proton, μpGEp/GMp, has been measured for elastic electron-proton scattering with polarized beam and target up to four-momentum transfer squared Q2=5.66(GeV/c)2 using double spin asymmetry for target spin orientation aligned nearly perpendicular to the beam momentum direction. This measurement of μpGEp/GMp agrees with the Q2 dependence of previous recoil polarization data and reconfirms the discrepancy at high Q2 between the Rosenbluth and the polarization-transfer method with a different measurement technique and systematic uncertainties uncorrelated to those of the recoil-polarization measurements. The form factor ratio at Q2=2.06(GeV/c)2 has been measured as μpGEp/GMp=0.720±0.176stat±0.039sys, which is in agreement with an earlier measurement using the polarized target technique at similar kinematics. The form factor ratio at Q2=5.66(GeV/c)2 has been determined as μpGEp/GMp=0.244±0.353stat±0.013sys, which represents the highest Q2 measurement reached using double spin asymmetries with polarized target to date.