Closed

Hierarchical BIGANs

BIGAN is a GAN that uses an auto-encoder as the discriminator. with a simple yet robust architecture, standard training procedure with fast and stable convergence.

In this work, we propose a hierarchical BIGANs architecture, because GAN has a perceptual loss function (THE DISCRIMINATOR). The discriminator itself has binary cross entropy as its loss function. Which cannot take perceptual information in the account. To mitigate the effect of the BCE loss we propose Hierarchical BIGANs.

Project Goals

1. What type of discriminator is best?

a. Can we use CNN as the discriminator?

b. Does the discriminator have to be an auto-encoder?

c. What impact would using other varieties of auto-encoders such Variational Auto Encoders[11] (VAEs) have?

2. What latent space size is best for a dataset and can it be used in a Semi supervised settings?

3. When should noise be added to the input and how much?

4. How can we Measure convergence, control distributional diversity and maintain the equilibrium between the discriminator and the generator.

5. Can Classification be done with images with random dimension/composition?

6. Can symbolic features like edge detection be combined with the original data, so that the performance of translation can be further improved.

7. Can we proposing a paradigm of modular learning which conducts continuous recursive self-improvement with regards to the utility function (reward system). We can view this as second (and higher) order optimization.

Habilidades: Image Processing, Mathlab y Mathematica, Neural Networks, Python, Arquitectura de software

Ver más: flash random images, random images within flash script, create flash header random images, bigan tensorflow, adversarially learned inference github, svhn gan, dumoulin adversarially learned inference, inferencing in adversarial generative models, hierarchical adversarially learned inference, adversarially learned inference tensorflow, generative adversarial network inference, web page header random images, flash header random images, random images samples flash, fullscreen random images flash, creating random images fading within flash movie, flash random images effects, random images text, random images flash, fade fade random images flash

Información del empleador:
( 2 comentarios ) Montreal, Canada

Nº del proyecto: #17587844

6 freelancers están ofertando el promedio de $116 para este trabajo

$155 CAD en 3 días
(22 comentarios)
5.8
dinhfreedom

Hello, there, I am very happy to put my bid on your project. I am an expert of algorithm and senior software developer, so I am interested in and confident to do this project. I hope to discuss everything of the pro Más

$150 CAD en 3 días
(55 comentarios)
6.1
williams115

Hi, there - My name is Phong. I have read your job description and I am very interested in this project because I have good experience with algorithm and programming. I would love to speak with you further about taki Más

$150 CAD en 3 días
(15 comentarios)
4.9
olaideejiwole

I have high proficiency in MATLAB & SIMULINK, mathematical modelling and analysis, Mathematics, Python,Algorithm, C programming, Machine learning analysis. Once you share the project, I will have a look & confirm if I Más

$30 CAD en 1 día
(13 comentarios)
4.1
rizwanmpec

Expert in Matlab, Simulink, Signal Processing, Image Processing Artificial Intelligence. Excellent in Electrical Engg. Face Detection, IRIS Detection, FingerPrint, Watermarking, Video processing. ANN, Fuzzy, GA, PSO, Más

$155 CAD en 3 días
(5 comentarios)
3.5
paulminu14

Dear Hiring Manager, I have worked with various Electrical & Electronic support companies as an Engineer which develop my skills in awarding with excellent, effective & efficient results as per my jobs responsibility. Más

$56 CAD en 3 días
(0 comentarios)
0.0