1: \begin{abstract}

2: Finding all linear inequalities for entropies remains an important open question in the information theory. For a long time the only known inequalities for entropies of tuples of random variables were Shannon inequalities (submodularity). Only in 1998 Zhang and Yeung found the first inequality that cannot be represented as a convex combination of Shannon inequalities, and several other non-Shannon inequalities were found soon after that. It turned out that the class of linear inequalities for entropies is rather fundamental, since the same class can be equivalently defined in terms of subgroup sizes or projections of multidimensional sets (Chan, Yeung, Romashchenko, Shen, Vereshchagin). The non-Shannon ineqialities have interesting applications (e.g., for secret sharing lower bounds). Still the class of linear equalities for entropies is not well understood (and some negative results are known, e.g., this class is not finitely axiomatizable).

3:  

4: This class appears also in algorithmic information theory: the same linear inequalities are true for Shannon entropies of tuples of random variables and Kolmogorov complexities of tuples of strings (Hammer et al., 1997). This parallelism started with the Kolmogorov--Levin formula (1968) for the complexity of pairs of strings with logarithmic precision. Longpr\'e (1986) proved a version of this formula for space-bounded complexities.

5: 

6: In this paper we prove an improved version of Longpr\'e's result with a tighter space bound, using Sipser's trick (1980). Then, using this space bound, we show that \emph{every linear inequality that is true for complexities or entropies, is also true for space-bounded Kolmogorov complexities with a polynomial space overhead}, thus extending the parallelism to the space-bounded algorithmic information theory.

7: \end{abstract}