-
Let's Have Both! Optimal List-Recoverability via Alphabet Permutation Codes
Authors:
Sergey Komech,
Jonathan Mosheiff
Abstract:
We introduce alphabet-permutation (AP) codes, a new family of error-correcting codes defined by iteratively applying random coordinate-wise permutations to a fixed initial word. A special case recovers random additive codes and random binary linear codes, where each permutation corresponds to an additive shift over a finite field.
We show that when these permutations are drawn from a suitably ``…
▽ More
We introduce alphabet-permutation (AP) codes, a new family of error-correcting codes defined by iteratively applying random coordinate-wise permutations to a fixed initial word. A special case recovers random additive codes and random binary linear codes, where each permutation corresponds to an additive shift over a finite field.
We show that when these permutations are drawn from a suitably ``mixing'' distribution, the resulting code is almost surely list-recoverable with list size proportional to the inverse of the gap to capacity. Compared to any linear code, our construction achieves exponentially smaller list sizes at the same rate. Previously, only fully random codes were known to attain such parameters, requiring exponentially many random bits and offering no structure. In contrast, AP codes are structured and require only polynomially many random bits -- providing the first such construction to match the list-recovery guarantees of random codes.
△ Less
Submitted 8 May, 2025; v1 submitted 9 February, 2025;
originally announced February 2025.
-
Choosing optimal parameters for a distributed multi-constrained QoS routing
Authors:
Sergey Komech,
Andrey Kupavskii,
Alexei Vezolainen
Abstract:
We consider several basic questions on distributed routing in directed graphs with multiple additive costs, or metrics, and multiple constraints. Distributed routing in this sense is used in several protocols, such as IS-IS and OSPF. A practical approach to the multi-constraint routing problem is to, first, combine the metrics into a single `composite' metric, and then apply one-to-all shortest pa…
▽ More
We consider several basic questions on distributed routing in directed graphs with multiple additive costs, or metrics, and multiple constraints. Distributed routing in this sense is used in several protocols, such as IS-IS and OSPF. A practical approach to the multi-constraint routing problem is to, first, combine the metrics into a single `composite' metric, and then apply one-to-all shortest path algorithms, e.g. Dijkstra, in order to find shortest path trees. We show that, in general, even if a feasible path exists and is known for every source and destination pair, it is impossible to guarantee a distributed routing under several constraints. We also study the question of choosing the optimal `composite' metric. We show that under certain mathematical assumptions we can efficiently find a convex combination of several metrics that maximizes the number of discovered feasible paths. Sometimes it can be done analytically, and is in general possible using what we call a 'smart iterative approach'. We illustrate these findings by extensive experiments on several typical network topologies.
△ Less
Submitted 11 October, 2023;
originally announced October 2023.
-
Modelling axon growing using CTRW
Authors:
Elena Zhizhina,
Sergey Komech,
Xavier Descombes
Abstract:
The main goal of this study is to propose a mathematical model describing paths of the axon growth cones and differences in the behavior of normal and mutant axons. We introduce a probabilistic model for axon growing, such that each family of axons is described as an ensemble of trajectories of a continuous time random walk (CTRW) model under different parameters in the case of normal and mutant a…
▽ More
The main goal of this study is to propose a mathematical model describing paths of the axon growth cones and differences in the behavior of normal and mutant axons. We introduce a probabilistic model for axon growing, such that each family of axons is described as an ensemble of trajectories of a continuous time random walk (CTRW) model under different parameters in the case of normal and mutant axons. We describe different regimes in the model and conclude how the behavior of axons depends on the parameters of the model. Biological observations of the axonal growth process say us that the guiding development of axons to their targets is operated by chemical signals from the cellular environment. To simulate this control mechanism we propose the CTRW model, where a random waiting time reflects a reaction time of the growth cones on the neighboring chemical environment.
△ Less
Submitted 5 December, 2015;
originally announced December 2015.
-
Shape Characterization via Boundary Distortion
Authors:
Xavier Descombes,
Serguei Komech
Abstract:
In this paper, we derive new shape descriptors based on a directional characterization. The main idea is to study the behavior of the shape neighborhood under family of transformations. We obtain a description invariant with respect to rotation, reflection, translation and scaling. A well-defined metric is then proposed on the associated feature space. We show the continuity of this metric. Some r…
▽ More
In this paper, we derive new shape descriptors based on a directional characterization. The main idea is to study the behavior of the shape neighborhood under family of transformations. We obtain a description invariant with respect to rotation, reflection, translation and scaling. A well-defined metric is then proposed on the associated feature space. We show the continuity of this metric. Some results on shape retrieval are provided on two databases to show the accuracy of the proposed shape metric.
△ Less
Submitted 24 February, 2013;
originally announced February 2013.
-
Entropy, Lyapunov exponents and the volume growth of boundary distortion under the action of dynamical systems
Authors:
B. M. Gurevich,
S. A. Komech
Abstract:
We study the connection between the Lyapunov exponents and the volume growth of boundary distortion of regions in the phase space of the dynamical system.
We study the connection between the Lyapunov exponents and the volume growth of boundary distortion of regions in the phase space of the dynamical system.
△ Less
Submitted 1 April, 2012;
originally announced April 2012.
