Quantum Bootstrap Aggregation
نویسندگان
چکیده
We set out a strategy for quantizing attribute bootstrap aggregation to enable variance-resilient quantum machine learning. To do so, we utilise the linear decomposability of decision boundary parameters in the Rebentrost et al. Support Vector Machine to guarantee that stochastic measurement of the output quantum state will give rise to an ensemble decision without destroying the superposition over projective feature subsets induced within the chosen SVM implementation. We achieve a linear performance advantage, O(d), in addition to the existing O(log(n)) advantages of quantization as applied to Support Vector Machines. The approach extends to any form of quantum learning giving rise to linear decision boundaries.
منابع مشابه
The CoIL Challenge: an Application of Classification Trees with Bootstrap Aggregation
A classification approach involving data tuning, followed by the use of a binary classification tree, augmented by bootstrap aggregation, yielded a ‘hit’ score of 110.
متن کاملOn the stability of support vector machines for face detection
In this paper we study the stability of support vector machines in face detection by decomposing their average prediction error into the bias, variance, and aggregation effect terms. Such an analysis indicates whether bagging, a method for generating multiple versions of a classifier from bootstrap samples of a training set, and combining their outcomes by majority voting, is expected to improv...
متن کاملBagging Is a Small-Data-Set Phenomenon
Bagging forms a committee of classijiers by bootstrap aggregation of training sets from a pool of training data. A simple alternative to bagging is to partition the data into disjoint subsets. Experiments on various datasets show that, given the same size partitions and bags, disjoint partitions result in betterperformance than bootstrap aggregates (bags). Many applications (e.g., protein struc...
متن کاملRoot Systems and Boundary Bootstrap
The principle of boundary bootstrap plays a significant role in the algebraic study of the purely elastic boundary reflection matrix Ka(θ) for integrable quantum field theory defined on a space-time with a boundary. However, general structure of that principle in the form as was originally introduced by Fring and Köberle has remained unclear. In terms of a new matrix Ja(θ) = √ Ka(θ)/Kā(iπ + θ),...
متن کاملFulfillment of the Strong Bootstrap Condition
The self-consistency of the assumption of Reggeized form of the production amplitudes in multi-Regge kinematics, which are used in the derivation of the BFKL equation, leads to strong bootstrap conditions. The fulfillment of these conditions opens the way to a rigorous proof of the BFKL equation in the next-to-leading approximation. The strong bootstrap condition for the kernel of the BFKL equa...
متن کامل