PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Share a file Manage my documents Convert Recover PDF Search Help Contact



DIVA poster final final (2) .pdf


Original filename: DIVA poster final final (2).pdf
Title: PowerPoint Presentation
Author: CanterburyMedia

This PDF 1.5 document has been generated by Microsoft® PowerPoint® 2013, and has been sent on pdf-archive.com on 24/01/2018 at 15:52, from IP address 207.38.x.x. The current document download page has been viewed 413 times.
File size: 574 KB (1 page).
Privacy: public file




Download original PDF file









Document preview


Evolving DIVA: Applying Evolution to a Model of Classification
Rafi Schulman
Advisor: Dr. Kenneth Kurtz

What is DIVA?
DIVA= DIVergent Autoencoder
• Model of Psychological Classification
• Machine Learning Approach

Evolving DIVA

Designing an evolutionary process to facilitate
classification learning in DIVA.

Future Directions
1. Learning on real-life data sets.

After generating an initial population of DIVA networks:
1. Evaluate each network’s performance on a
classification task and assign a fitness value to that
network
2. Replicate networks with higher fitness, remove
networks with lower fitness

2. Identifying successful “Traits” within
DIVA’s weights

3. Create “mutations” in some networks by altering
their weights

“Categories are represented as
coordinated statistical models of the
properties of the members” 1
2
backpropagation ,

DIVA Learns through
a supervised algorithm to optimize
weights through error-driven learning
by gradient descent. However,
backpropagation is sometimes slow,
too computationally demanding with
larger nets, and can get stuck in local
minima.

Cognitive Science to Inform
Machine Learning Development
Machine learning is a growing field
with expanding application.
Psychological models and theories
can be an essential tool to this
research field, which is predominantly
made up of computer scientists
RESEARCH POSTER PRESENTATION DESIGN © 2015

www.PosterPresentations.com

4. Repeat the evolutionary cycle

Method
A population of 10 individual networks
was randomly generated. Fitness was
assigned based on the summed
classification probabilities of the correct
category in 8 examples of a four-variable
categorization problem. At each
generation, the five least fit individuals
were removed from the population. The
most fit individual was replicated twice,
and the second and third most fit
individuals were replicated once. Each
replicated network was mutated by
random variation added to each weight in
it’s matrix. Additionally, a new random
individual was introduced at each
generation. This was repeated for 10 trials
of 50 generations. The software was
written using Python 3.3 with the NumPy
matrix multiplication package.

Observations/Results
Upon initialization of the
population, summed
classification probabilities
ranged from 3.8-4.2. In
most trials, the summed
classification probabilities
in the most fit individual
Stimulus and categories used
rose to 5.2 by the twentieth
generation, and within a few generations
had spread to the whole population. However,
in no trial did this statistic rise above 5.2. I
believe that this is because random mutations
stopped improving network performance.
Additionally, it was found that the randomly
introduced individuals at each generation
sometimes had high relative fitness and
consequently played a large part in overall
population dynamics.

3. “Sexual Reproduction”
4. Integrating individual learning with
population learning: Backprop and
Evolution for local and global
maximums.

Acknowledgements/References
Professor Ken Kurtz
Doctoral Candidate Nolan Conaway
The EvoS Program at Binghamton
University of California Irvine Machine Learning
Data Repository
1: Kurtz, K.J. (2007) The divergent autoencoder
(DIVA) Model of category Learning. Psychonomic
Bulletin and Review, 14(4), 560-567.
2: Rumelhart D.E., Hinton G.E., Williams R.J. (1986).
Learning Representations by back-propagating
errors. Nature, 323. 533-536


Document preview DIVA poster final final (2).pdf - page 1/1

Related documents


diva poster final final 2
4n13 ijaet0313474 revised
ihmsc 2017 132
projet hopfield
resume
saksham singhal resume


Related keywords