Quick Navigation
Topics
Trapped Ion Quantum Computing
Quantum Machine Learning
A streamlined quantum algorithm for topological data analysis with exponentially fewer qubits
arXiv
Authors: Sam McArdle, András Gilyén, Mario Berta
Year
2022
Paper ID
58942
Status
Preprint
Abstract Read
~2 min
Abstract Words
168
Citations
N/A
Abstract
Topological invariants of a dataset, such as the number of holes that survive from one length scale to another (persistent Betti numbers) can be used to analyze and classify data in machine learning applications. We present an improved quantum algorithm for computing persistent Betti numbers, and provide an end-to-end complexity analysis. Our approach provides large polynomial time improvements, and an exponential space saving, over existing quantum algorithms. Subject to gap dependencies, our algorithm obtains an almost quintic speedup in the number of datapoints over previously known rigorous classical algorithms for computing the persistent Betti numbers to constant additive error - the salient task for applications. However, we also introduce a quantum-inspired classical power method with provable scaling only quadratically worse than the quantum algorithm. This gives a provable classical algorithm with scaling comparable to existing classical heuristics. We discuss whether quantum algorithms can achieve an exponential speedup for tasks of practical interest, as claimed previously. We conclude that there is currently no evidence for this being the case.
Why This Paper Matters
- This paper contributes to the Quantum Machine Learning research area in the Quantum Articles archive.
- It adds a 2022 reference point for readers tracking recent quantum research.
- Topological invariants of a dataset, such as the number of holes that survive from one length scale to another (persistent Betti numbers) can be used to analyze and classify...
Paper Tools
Become a member to use research tools
Sign in to open papers, visit source links, share, cite, compare, copy DOI links, request category corrections, and build your reading list.
Show Paper arXiv Publisher Share
Cite This Paper
Copy URL
Compare
Copy DOI Add to Reading List
Category Correction Request
Category Correction Request
Help us improve classification quality by proposing a better category. Every request is reviewed by an admin.
Sign in to submit a category correction request for this paper.
Log In to SubmitReferences & Citation Signals
Community Reactions
Quick sentiment from readers on this paper.
Score:
0
Likes: 0
Dislikes: 0
Sign in to react to this paper.
Discussion & Reviews (Moderated)
Average Rating: 0.0 / 5 (0 ratings)
No written reviews yet.