Danny Dubé

A hybrid approach to vehicle routing using neural networks and genetic algorithms

A competitive neural network model and a genetic algorithm are used to improve the initialization and construction phase of a parallel insertion heuristic for the vehicle routing problem with time windows. The neural network identifies seed customers that are distributed over the entire geographic area during the initialization phase, while the genetic algorithm finds good parameter settings in the route construction phase that follows. Computational results on a standard set of problems are also reported.

Lossless Data Compression via Substring Enumeration

We present a technique that compresses a string

Efficiently building a parse tree from a regular expression

w

The universality and linearity of compression by substring enumeration

by enumerating all the substrings of

Improving a vehicle routing heuristic through genetic search

w

Using synchronization bits to boost compression by substring enumeration

. The substrings are enumerated from the shortest to the longest and in lexicographic order. Compression is obtained from the fact that the set of the substrings of a particular length gives a lot of information about the substrings that are one bit longer. A linear-time, linear-space algorithm is presented. Experimental results show that the compression efficiency comes close to that of the best PPM variants. Other compression techniques are compared to ours.

Improving LZ77 bit recycling using all matches

Abstract. We show in this paper that parsing with regular expressions instead of context-free grammars, when it is possible, is desirable. We present efficient algorithms for performing different tasks that concern parsing: producing the external representation and the internal representation of parse trees; producing all possible parse trees or a single one. Each of our algorithms to produce a parse tree from an input string has an optimal time complexity, linear with the length of the string. Moreover, ambiguous regular expressions can be used.

An extended proof-carrying code framework for security enforcement

A new lossless data compression technique called compression by substring enumeration (CSE) has recently been introduced. Two conjectures have been stated in the original paper and they have not been proved there nor in subsequent papers on CSE. The first conjecture says that CSE is universal for Markovian sources, provided an appropriate predictor is devised. The second one says that CSE has a linear complexity both in time and in space. In this paper, we present an appropriate predictor and demonstrate that CSE indeed becomes universal for any order-k Markovian source. Finally, we prove that the compacted substring tree on which CSEs linear complexity depends effectively has linear size.

On the Use of Stronger Synchronization to Boost Compression by Substring Enumeration

A genetic algorithm is applied to the search of good parameter settings for a vehicle routing heuristic. The parameter settings identified by the genetic search allow the insertion heuristic to generate solutions that are much better than the solutions previously reported on a standard set of routing problems.<<ETX>>

Constructing optimal whole-bit recycling codes

A new lossless data compression technique called compression via substring enumeration (CSE) has recently been introduced. It has been observed that CSE achieves lower performance on binary data. An hypothesis has been formulated that suggests that CSE loses track of the position of the bits relative to the byte boundaries more easily in binary data and that this confusion incurs a penalty for CSE. This paper questions the validity of the hypothesis and proposes a simple technique to reduce the penalty, in case the hypothesis is correct. The technique consists in adding a preprocessing step that inserts synchronization bits in the data in order to boost the performance of CSE. Experiments provide strong evidence that the formulated hypothesis is true and they demonstrate the effectiveness of the use of synchronization bits.