IMPROVING THE ARCHITECTURE AND INTERPRETABILITY OF DEEP NEURAL NETWORKS: AN EVOLUTIONARY COMPUTING APPROACH

BASIC INFORMATION

Ph.D. Student: Antonio R. Moya Martín-Castaño
Advisor: Sebastián Ventura
Started on: October 2019
Keywords: Evolutionary Algortihms, Neural Networks

THESIS PROPOSAL

Many advances have been made in recent years in the field of artificial intelligence. Within these advances, the importance of a growing paradigm such as deep learning stands out, in which models inspired by the structure of the neuron networks of the human nervous system are developed to perform computationally costly tasks such as image recognition, speech synthesis, etc.

For proper learning in these deep models, it is necessary to properly adjust the architecture and hyper-parameters of these models, a task that is computationally impossible to do by brute force. For the optimisation of architecture and hyper-parameters, numerous techniques have already been developed. These include, for example, random search, grid search or bayesian optimisation techniques. In our proposal, we focus on other techniques also used for this and other areas: evolutionary algorithms. We can find in the literature numerous cases in which the use of these algorithms in optimisation problems leads to great results. From this type of algorithms we highlight their capacity to balance exploitation and exploration during the search process.

Despite reducing the time required by a brute-force technique, optimisation techniques (even more so in the case of evolutionary algorithms) generally have to deal with very high computing times. An evolutionary algorithm is therefore proposed in which the process of searching for the best solution is guided by partial evaluations of the deep learning models, obtaining an approximate result in each of these evaluations that indicates how good could be each model, so that the time spent in general by the evolutionary algorithm is notably reduced.

For this task of prediction from model evaluations, probabilistic models, that are defined by a combination of functions that explain the learning curve, are used.

The partial objectives are the following:

  • Define an evolutionary algorithm that allows the optimisation of architecture and hyper-parameters of deep learning models used for different tasks, such as image classification, natural language processing or human activity recognition.
  • Define probabilistic models that allow adequate prediction of how a model learns from a partial evaluation of this model.
  • Use these probabilistic models to reduce evaluation times during the evolutionary algorithm.

FUNDS

The development of this thesis is being supported by:

  • Spanish Ministry of Education, Culture and Sports under the FPU program (FPU18/06307).

At the ages of fifty four fifty can advise you that I'onal ended up lucky not to have wanted the product sooner, nevertheless loosing your partner of 25yrs 2010 became a curve which modified me personally for a long time. Now there came out a place exactly where click to read fifty had visit to have my tastes fulfilled only to find out this plumbing related desired just a little poke. So I questioned my own Computer system doc intended for a little something with tiny facet is affecting. He or she provided the particular recommended you read Cialis regular 5mg. 1st working day fifty had 5mg without any help to discover more help in the event that t discovered virtually any difference considering the woman never was planning determine, which'azines our system and also l'meters being dedicated to the idea. Regardless, these materials Operates, along with is useful. And click here then up coming night time with your ex m took 10mg at 8pm, and the rest is heritage. Through 13:double zero fifty manufactured my own move, but it appeared to be the most element to live in place until finally the sunlight came up upward , 100%Pleased :)When i't thirty-two as well as gone pretty much 1,5 years without intercourse. I had created pop over to this website plenty of anxiousness related penile erection challenges. And hop over to here then We found this specific great which woman My spouse and i started courting, along with first 2 times us all sex didn'to determine which properly, and i also appeared to be worried about just what exactly your woman considered this matter. We obtained braveness to visit and request cialis approved from the health practitioner. When i had taken 10mg product and it labored perfectly. I could truthfully continue on having sex many times a day without the problems. Merely bad thing is a smallish frustration. For me this can be truly a wonder drug.