Online Learning Algorithms

2020 ◽  
Vol 8 (1) ◽  
Author(s):  
Nicolò Cesa-Bianchi ◽  
Francesco Orabona
Keyword(s):  
Online Learning ◽  
Convex Optimization ◽  
Online Algorithms ◽  
Analysis Of Algorithms ◽  
Annual Review ◽  
Publication Date ◽  
Sequential Data ◽  
Theoretic Approach ◽  
Mirror Descent ◽  
Information Theoretic Approach

Online learning is a framework for the design and analysis of algorithms that build predictive models by processing data one at the time. Besides being computationally efficient, online algorithms enjoy theoretical performance guarantees that do not rely on statistical assumptions on the data source. In this review, we describe some of the most important algorithmic ideas behind online learning and explain the main mathematical tools for their analysis. Our reference framework is online convex optimization, a sequential version of convex optimization within which most online algorithms are formulated. More specifically, we provide an in-depth description of online mirror descent and follow the regularized leader, two of the most fundamental algorithms in online learning. As the tuning of parameters is a typically difficult task in sequential data analysis, in the last part of the review we focus on coin-betting, an information-theoretic approach to the design of parameter-free online algorithms with good theoretical guarantees. Expected final online publication date for the Annual Review of Statistics, Volume 8 is March 8, 2021. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

scholarly journals Effects of Causes and Causes of Effects

2021 ◽  
Vol 9 (1) ◽  
Author(s):  
A. Philip Dawid ◽  
Monica Musio
Keyword(s):  
Individual Case ◽  
Causal Link ◽  
Structural Equations ◽  
Potential Outcomes ◽  
Annual Review ◽  
Publication Date ◽  
Theoretic Approach ◽  
Additional Structure ◽  
Decision Theoretic Approach ◽  
Application Volume

We describe and contrast two distinct problem areas for statistical causality: studying the likely effects of an intervention (effects of causes) and studying whether there is a causal link between the observed exposure and outcome in an individual case (causes of effects). For each of these, we introduce and compare various formal frameworks that have been proposed for that purpose, including the decision-theoretic approach, structural equations, structural and stochastic causal models, and potential outcomes. We argue that counterfactual concepts are unnecessary for studying effects of causes but are needed for analyzing causes of effects. They are, however, subject to a degree of arbitrariness, which can be reduced, though not in general eliminated, by taking account of additional structure in the problem. Expected final online publication date for the Annual Review of Statistics and Its Application, Volume 9 is March 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Value of Information Analysis in Models to Inform Health Policy

2021 ◽  
Vol 9 (1) ◽  
Author(s):  
Christopher H. Jackson ◽  
Gianluca Baio ◽  
Anna Heath ◽  
Mark Strong ◽  
Nicky J. Welton ◽  
...  
Keyword(s):  
Value Of Information ◽  
Evidence Synthesis ◽  
Annual Review ◽  
Publication Date ◽  
Theoretic Approach ◽  
Information Analysis ◽  
Decision Theoretic Approach ◽  
Expected Benefits ◽  
Different Sources ◽  
Broad Overview

Value of information (VoI) is a decision-theoretic approach to estimating the expected benefits from collecting further information of different kinds, in scientific problems based on combining one or more sources of data. VoI methods can assess the sensitivity of models to different sources of uncertainty and help to set priorities for further data collection. They have been widely applied in healthcare policy making, but the ideas are general to a range of evidence synthesis and decision problems. This article gives a broad overview of VoI methods, explaining the principles behind them, the range of problems that can be tackled with them, and how they can be implemented, and discusses the ongoing challenges in the area. Expected final online publication date for the Annual Review of Statistics, Volume 9 is March 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

scholarly journals Environmental factors affecting the wintering raptor community in Armenia, Southern Caucasus

2021 ◽  
Author(s):  
Gianpasquale Chiatante ◽  
Michele Panuccio
Keyword(s):  
General Pattern ◽  
Abundant Species ◽  
Theoretic Approach ◽  
Factors Affecting ◽  
Milvus Migrans ◽  
Gyps Fulvus ◽  
Black Kite ◽  
Southern Caucasus ◽  
And Migration ◽  
Information Theoretic Approach

AbstractThe species–habitat relationships can change during the year because of the seasonality of resources. Therefore, the investigation of habitat use by animals in each season plays a fundamental role in their conservation. The main aim of this research was to investigate the raptor community that spends the winter in Armenia, southern Caucasus, and to explore its relationship with environmental features, such as land use and topography. During January 2012, we collected data by carrying out 15 roadside counts along which we calculated three community parameters: the relative abundance, the species richness, and the species diversity. Then, we carried out a multiple linear regression with the Information-Theoretic Approach, to explain the relationship between the parameters and environmental variables. Besides, we computed a Canonical Correspondence Analysis (CCA) between the species and the environment around their observations. As a general pattern, the community was associated with permanent crops, maybe because of their heterogeneity, which in turn allows them to support higher densities of prey during the winter. The most abundant species was the Black Kite (Milvus migrans), followed by the Common Kestrel (Falco tinnunculus) and the Griffon Vulture (Gyps fulvus). To our knowledge, this is one of the first studies investigating the wintering raptor community in the Caucasus, with raptors generally studied in this area during the breeding season and migration.

scholarly journals Causality and Information Transfer Between the Solar Wind and the Magnetosphere–Ionosphere System

Entropy ◽  
2021 ◽  
Vol 23 (4) ◽  
pp. 390
Author(s):  
Pouya Manshour ◽  
Georgios Balasis ◽  
Giuseppe Consolini ◽  
Constantinos Papadimitriou ◽  
Milan Paluš
Keyword(s):  
Solar Wind ◽  
Information Transfer ◽  
Linear Time ◽  
Vertical Component ◽  
Causal Link ◽  
Theoretic Approach ◽  
Geomagnetic Disturbances ◽  
Magnetospheric Substorms ◽  
Solar Wind Parameters ◽  
Information Theoretic Approach

An information-theoretic approach for detecting causality and information transfer is used to identify interactions of solar activity and interplanetary medium conditions with the Earth’s magnetosphere–ionosphere systems. A causal information transfer from the solar wind parameters to geomagnetic indices is detected. The vertical component of the interplanetary magnetic field (Bz) influences the auroral electrojet (AE) index with an information transfer delay of 10 min and the geomagnetic disturbances at mid-latitudes measured by the symmetric field in the H component (SYM-H) index with a delay of about 30 min. Using a properly conditioned causality measure, no causal link between AE and SYM-H, or between magnetospheric substorms and magnetic storms can be detected. The observed causal relations can be described as linear time-delayed information transfer.

Endogenous Retroviruses Drive Resistance and Promotion of Exogenous Retroviral Homologs

2020 ◽  
Vol 9 (1) ◽  
Author(s):  
Elliott S. Chiu ◽  
Sue VandeWoude
Keyword(s):  
Immune Modulation ◽  
Viral Infections ◽  
Viral Evolution ◽  
Endogenous Retroviruses ◽  
Annual Review ◽  
Publication Date ◽  
Biological Functions ◽  
Self Tolerance ◽  
Vertebrate Hosts ◽  
Insight Into

Endogenous retroviruses (ERVs) serve as markers of ancient viral infections and provide invaluable insight into host and viral evolution. ERVs have been exapted to assist in performing basic biological functions, including placentation, immune modulation, and oncogenesis. A subset of ERVs share high nucleotide similarity to circulating horizontally transmitted exogenous retrovirus (XRV) progenitors. In these cases, ERV–XRV interactions have been documented and include ( a) recombination to result in ERV–XRV chimeras, ( b) ERV induction of immune self-tolerance to XRV antigens, ( c) ERV antigen interference with XRV receptor binding, and ( d) interactions resulting in both enhancement and restriction of XRV infections. Whereas the mechanisms governing recombination and immune self-tolerance have been partially determined, enhancement and restriction of XRV infection are virus specific and only partially understood. This review summarizes interactions between six unique ERV–XRV pairs, highlighting important ERV biological functions and potential evolutionary histories in vertebrate hosts. Expected final online publication date for the Annual Review of Animal Biosciences, Volume 9 is February 16, 2021. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Sign in / Sign up

Export Citation Format

Share Document