AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

An Algorithm for Learning Switched Linear Dynamics from Data

Guillaume O Berger, Monal Narasimhamurthy, Kandai Watanabe, Morteza Lahijanian, Sriram Sankaranarayanan

NeurIPS 2022 · Learning theory and algorithms › Combinatorial and search optimization · approach: Predict-then-optimize learning · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use proposed gap-based algorithm instead of kLRSwitched linear ID · Benchmark 1, N=20one-step error: 0.34 -> 0.17Improved
use proposed algorithm instead of kLRSwitched linear ID · Benchmark 3, N=100one-step error: 0.48 -> 0.23Improved
use proposed algorithm instead of kLRSwitched linear ID · Benchmark 3, N=100training time (s): 0.41 -> 506.29Hurt
use proposed algorithm instead of kLRSwitched linear ID · Benchmark 2, N=500one-step error: 0.27 -> 0.06Improved
use proposed algorithm instead of kLRSwitched linear ID · Acrobotone-step error: 1.80 -> 0.005Improved
use proposed algorithm instead of kLRSwitched linear ID · Cartpoleone-step error: 0.04 -> 0.016Improved
use proposed algorithm instead of kLRSwitched linear ID · Acrobottraining time (s): 0.18 -> 7.56Hurt
increase data points N from 20 to 500Proposed approach · Benchmark 1one-step error: 0.17 -> 0.05Improved
increase data points N from 100 to 500Proposed approach · Benchmark 3training time (s): 506.29 -> 1184.16Hurt
increase N from 20 to 500k-Linear Regression · Benchmark 2one-step error: 0.29 -> 0.27No clear effect
replace MILP solver with proposed algorithm as N growsProposed approach vs MILP (Gurobi) · synthetic microbenchmarkscomputation timeImproved
vary state dimension n or number of modes mProposed vs MILP · synthetic microbenchmarkscomputation timeNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy