Probabilistic Bisection Converges Almost as Quickly as Stochastic Approximation
This work addresses a limitation in stochastic root-finding methods for applications where error probabilities vary, though it is incremental as it builds on existing probabilistic bisection algorithms.
The paper tackles the problem of stochastic root-finding when the assumption of fixed error probability in responses is relaxed, by extending the probabilistic bisection algorithm with a power-one test. The result shows that this extended algorithm converges at a rate arbitrarily close to, but slower than, the square root rate of stochastic approximation.
The probabilistic bisection algorithm (PBA) solves a class of stochastic root-finding problems in one dimension by successively updating a prior belief on the location of the root based on noisy responses to queries at chosen points. The responses indicate the direction of the root from the queried point, and are incorrect with a fixed probability. The fixed-probability assumption is problematic in applications, and so we extend the PBA to apply when this assumption is relaxed. The extension involves the use of a power-one test at each queried point. We explore the convergence behavior of the extended PBA, showing that it converges at a rate arbitrarily close to, but slower than, the canonical "square root" rate of stochastic approximation.