Boosting: Foundations and Algorithms

By A. Andrew. 2013

752 citations recorded by the source corpus.

Subfields: Optimization & Training, Statistical Learning Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Optimization & Training

In Statistical Learning Theory

Citations in the corpus

Machine Learning Methods That Economists Should Know About (2019, background): “See Schapire and Freund [2012] for a detailed discussion.”
Model Zoo: A Growing Brain That Learns Continually (2021, methodology): “Unlike AdaBoost which combines all the weak-learners using speciﬁc weights, we simply average the predictions of all models trained on each task.”
A Confidence-Based Approach for Balancing Fairness and Accuracy (2016, background): “In particular, RR computes the probability p for which, if members of S with label −1 under h are flipped by h′ to +1 randomly and independently with probability p, the bias of h′ is zero in expectation.”
Multiaccuracy: Black-Box Post-Processing for Fairness in Classification (2018, methodology): “The post-processing algorithm is an iterative procedure similar to boosting (Schapire and Freund 2012), that uses the multiplicative weights framework to improve suboptimal predictions identified by the auditor.”
Data Programming: Creating Large Training Sets, Quickly (2016, background): “Boosting is a well known procedure for combining the output of many “weak” classiﬁers to create a strong classiﬁer in a supervised setting [26].”
Balancing Between Forgetting and Acquisition in Incremental Subpopulation Learning (2022, methodology): “…the final classifier G ϕ t for model F t by linear addition: This is also inspired from boosting mechanism, but it is totally different from boosting since the α t here is for controlling the learning and forgetting, while in boosting we use it to progressively reduce the training error [29].”
Overfitting or perfect fitting? Risk bounds for classification and regression rules that interpolate (2018, background): “The literature on this subject is vast, ranging from classical parametric and non-parametric statistics [48, 49] to more recent machine learning methods, such as kernel machines [39], boosting [36], random forests [15], and deep neural networks [25].”
Explaining AdaBoost (2013, methodology): “To demonstrate this point, consider the following experiment from Schapire and Freund [30], which is similar in spirit to the work of Mease and Wyner [24] [37].”
Complexity theoretic limitations on learning halfspaces (2015, methodology): “The problem of learning halfspaces is as old as the field of machine learning, starting with the perceptron algorithm [56, 55], through the modern SVM [27, 61, 60, 28, 29] and AdaBoost [58, 57, 40].”
A machine-learning heuristic to improve gene score prediction of polygenic traits (2017, methodology): “∑ = − = d Z Z L f d f ( , ( )) ( ( )) (6) j j m j 1 2”
A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects (2022, methodology): “Hence, bootstrap methods such as random forest [51] and boosting methods such as AdaBoost [52] have been developed to achieve diversity in the ensemble.”
Re-evaluating evaluation (2018, background): “It takes an adversarial perspective and may not be the best approach to, say, constructing automated curricula – although boosting is a related approach that works well [81,82].”
Inferring Generative Model Structure with Static Analysis (2017, background): “Methods such as crowdsourcing [12, 17, 40], boosting[3, 33], co-training [6], and learning from noisy labels are some of the popular approaches that can combine various sources of weak supervision to assign noisy labels to data.”
Multi-Class Deep Boosting (2014, background): “, 2001a] (see [Schapire and Freund, 2012] for a more extended list of references), as discussed by Cortes et al.”
The Algorithmic Foundations of Differential Privacy (2014, background): “See Schapire [77] for an excellent survey of boosting, and the textbook “Boosting” by Freund and Schapire [79] for a thorough treatment.”
Functional Frank-Wolfe Boosting for General Loss Functions (2015, methodology): “It is well known that AdaBoost can be derived from functional gradient decent method using exponential loss [1, 27].”
In-Memory Computation of a Machine-Learning Classifier in a Standard 6T SRAM Array (2017, methodology): “Error-adaptive classifier boosting (EACB) [12] is an approach that extends from Adaptive Boosting (AdaBoost) [11].”