AI & ML
Block encoding of semiseparable matrices via their generators
arXiv cs.NA0 views
arXiv:2610.13228v1 Announce Type: cross Abstract: We present an explicit quantum circuit that constructs a block encoding of a generator-representable semiseparable (one-pair) matrix directly from its two generators. For an $N\times N$ semiseparable matrix with $N=2^{n}$, the circuit uses $2n+2$ ancilla qubits and has $O({\rm polylog} N)$ depth, and it realizes a $(2\sqrt{N},\,2n+2,\,2 \sqrt{2}\ \varepsilon)$-block encoding, where $\varepsilon$ is controlled by the accuracy of the state-preparation routines that load the generators. The building blocks of the circuit are cyclic shifts, swaps, and Hadamard gates, each admitting an efficient implementation through the binary decomposition of the relevant indices.
Source:https://arxiv.org/abs/2610.13228PreviousPolyconvex neural-network surrogates with learned paired directional features for anisotropic hyperelastic homogenizationNextSustainable Cloud-Fog-Edge Systems for Smart Buildings: Algorithmic Adaptation under Memory Constraints and Dynamic Orchestration Strategies
Related posts
Hybrid retrieval with reciprocal rank fusion: notes from a RAG prototype
0dev.to
It is in co-creating, together. As humans do.
0dev.to
C64 Basic Dungeon Crawler: Stairs, Levels and Levelling Up (Part 11)
0Hacker News
Anthropic's internal AI use, at retail prices, now exceeds its employee pay
0Hacker News
Launch HN: Momentic (YC W24) – Agents that catch bugs by using your app
0Hacker News

Behavioral Design Unpacked: Nudges, Funnels & Persuasion | UX Unpacked #3
0dev.to UX