Evolving a Deep Neural Network Training Time Estimator

in Communications in Computer and Information Science (2020, February)

We present a procedure for the design of a Deep Neural Net- work (DNN) that estimates the execution time for training a deep neural network per batch on GPU accelerators. The estimator is destined to be ... [more ▼]

We present a procedure for the design of a Deep Neural Net- work (DNN) that estimates the execution time for training a deep neural network per batch on GPU accelerators. The estimator is destined to be embedded in the scheduler of a shared GPU infrastructure, capable of providing estimated training times for a wide range of network architectures, when the user submits a training job. To this end, a very short and simple representation for a given DNN is chosen. In order to compensate for the limited degree of description of the basic network representation, a novel co-evolutionary approach is taken to fit the estimator. The training set for the estimator, i.e. DNNs, is evolved by an evolutionary algorithm that optimizes the accuracy of the estimator. In the process, the genetic algorithm evolves DNNs, generates Python-Keras programs and projects them onto the simple representation. The genetic operators are dynamic, they change with the estimator’s accuracy in order to balance accuracy with generalization. Results show that despite the low degree of information in the representation and the simple initial design for the predictor, co-evolving the training set performs better than near random generated population of DNNs. [less ▲]

Formal Security Analysis of Traditional and Electronic Exams

in Communications in Computer and Information Science (2015), 554

Nowadays, students can be assessed not only by means of pencil-and-paper tests but also by electronic exams which they take in examination centers or even from home. Electronic exams are appealing as they ... [more ▼]

Nowadays, students can be assessed not only by means of pencil-and-paper tests but also by electronic exams which they take in examination centers or even from home. Electronic exams are appealing as they can reach larger audiences, but they are exposed to new threats that can potentially ruin the whole exam business. These threats are amplified by two issues: the lack of understanding of what security means for electronic exams (except the old concern about students cheating), and the absence of tools to verify whether an exam process is secure. This paper addresses both issues by introducing a formal description of several fundamental authentication and privacy properties, and by establishing the first theoretical framework for an automatic analysis of exam security. It uses the applied π-calculus as a framework and ProVerif as a tool. Three exam protocols are checked in depth: two Internet exam protocols of recent design, and the pencil-and-paper exam used by the University of Grenoble. The analysis highlights several weaknesses. Some invalidate authentication and privacy even when all parties are honest; others show that security depends on the honesty of parties, an often unjustified assumption in modern exams. [less ▲]