I new developments in this area and applying the

I am deeply
interested in Machine Learning and Robotics specifically in its applications to
Vision and Self Learning. I have invested my time in researching new
developments in this area and applying the knowledge that I have acquired to
innovative use.

If Ray Kurzweil is
right in his book “Singularity is Near”, we are near the end of ‘Epoch 4’
(Technology) and entering into ‘Epoch 5’ (The Merger of Human Technology with
Human Intelligence). In the distant future, as Ray points out in his book –
advanced technology will transcend the human brain’s limitations of a mere
hundred trillion extremely slow connections.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

While Ray’s ideas
are interesting and exciting to think about, in the near future I foresee Artificial
Intelligence and Machine Learning being used as a tool, as mathematics is used,
for all the fields of science. Machine learning from big data allows systems to
learn subtle statistical regularities of the visual world, but the humans have
the ability to learn from very few examples. Today, Machine Learning
algorithms require large datasets to predict accurately. Acquiring of such
large datasets is difficult especially in fields like Healthcare etc. due to
Security Reasons or due to the fact that the data is unavailable. A new type of
Neural Networks called ‘Generative Adversarial Networks’ has emerged from
recent research developments which seems to solve the above problem. I would
like to study and explore how ‘Generative Adversarial Networks’ can help to
solve the above-mentioned issue. Another problem that is being faced is the
process of manually designing machine learning models because the search space
of all possible models can be combinatorically large and it often takes a
significant amount of time and experimentation even for an expert. Thus, there
is a need for self-learning. I would like to research on self- learning i.e.
reinforcement learning which learns to predict accurately from its previous
predictions and implement this to predict even better learning models or produce
new types of neural networks helping non-experts and thus allowing machine learning
to have a greater impact to everyone.