A bucket-and-boost estimator using least squares and a geometric median recovers the first T Markov parameters of an LTI system under heavy-tailed noise with O(sqrt(p log(1/delta)/N)) error and O((mT)^2 kappa log(1/delta)) samples.
Asymptotic properties o f general autoregressive models and strong consistency of least-squares estimates of their parameters
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SY 1years
2025 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Boosting-Enabled Robust System Identification of Partially Observed LTI Systems Under Heavy-Tailed Noise
A bucket-and-boost estimator using least squares and a geometric median recovers the first T Markov parameters of an LTI system under heavy-tailed noise with O(sqrt(p log(1/delta)/N)) error and O((mT)^2 kappa log(1/delta)) samples.