Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Between SC and LOGDCFL: Families of Languages Accepted by Logarithmic-Space Deterministic Auxiliary Depth-k Storage Automata (2203.10193v2)

Published 18 Mar 2022 in cs.FL and cs.CC

Abstract: The closure of deterministic context-free languages under logarithmic-space many-one reductions ($\mathrm{L}$-m-reductions), known as LOGDCFL, has been studied in depth from an aspect of parallel computability because it is nicely situated between $\mathrm{L}$ and $\mathrm{AC}{1}\cap\mathrm{SC}2$. By replacing a memory device from pushdown stacks with access-controlled storage tapes, we introduce a computational model of one-way deterministic depth-$k$ storage automata ($k$-sda's) whose tape cells are freely modified during the first $k$ accesses and then become blank forever. These $k$-sda's naturally induce the language family $k\mathrm{SDA}$. Similarly to $\mathrm{LOGDCFL}$, we study the closure $\mathrm{LOG}k\mathrm{SDA}$ of all languages in $k\mathrm{SDA}$ under $\mathrm{L}$-m-reductions. We demonstrate that $\mathrm{DCFL}\subseteq k\mathrm{SDA}\subseteq \mathrm{SC}k$ by significantly extending Cook's early result (1979) of $\mathrm{DCFL}\subseteq \mathrm{SC}2$. The entire hierarch of $\mathrm{LOG}k\mathrm{SDA}$ for all $k\geq1$ therefore lies between $\mathrm{LOGDCFL}$ and $\mathrm{SC}$. As an immediate consequence, we obtain the same simulation bounds for Hibbard's limited automata. We further characterize $\mathrm{LOG}k\mathrm{SDA}$ in terms of a new machine model, called logarithmic-space deterministic auxiliary depth-$k$ storage automata that run in polynomial time. These machines are as powerful as a polynomial-time two-way multi-head deterministic depth-$k$ storage automata. We also provide a ``generic'' $\mathrm{LOG}k\mathrm{SDA}$-complete language under $\mathrm{L}$-m-reductions by constructing a two-way universal simulator working for all $k$-sda's.

Citations (1)

Summary

We haven't generated a summary for this paper yet.