Deep learning involves learning through layers which allows a computer to build a hierarchy of complex concepts out of simpler concepts. Just like Product Management, the objective of Deep Learning is to solve ‘intuitive’ problems i.e. problems characterized by High dimensionality and no rules.
In this talk, Moriya discussed with us how deep is the future of IoT, how is it changing the way we create products and what will be its implications.
6. Data as itself is not a fortune, but a problem.
It becomes a fortune, when you drive it to
Conclusions ---->Value
7. So: What is it a SMART product?
Or: how can a product bring more value than its sensor?
8. It always starts with good people and their innate
passion to solve difficult problems!
The technology is just a tool to fulfil their quest.
9. The recent surge of data such as images, text, speech
enabled by cellular phones and mobile devices has
created a need to understand this complex data that was
not machine understandable and searchable.
10. While the initial technology challenge in harnessing IoT
is an infrastructural upgrade to address the data
storage, integration, and analytic requirements, the end
goal is to generate meaningful business insights from
the ocean of data that can translate to strategic business
advantages.
11. The idea is that: if like humans, Computers were to
gather knowledge from experience, it avoids the need
for human operators to formally specify all of the
knowledge that the computer needs to solve a problem.
Deep Learning is used to address intuitive applications
with high dimensionality.
12. Deep learning is often thought of as a set of algorithms
that ‘mimics the brain’. A more accurate description
would be an algorithm that ‘learns in layers’. Deep
learning involves learning through layers which allows a
computer to build a hierarchy of complex concepts out
of simpler concepts.
Deep learning algorithms apply to many areas including
Computer Vision, Image recognition, pattern recognition,
speech recognition, behaviour recognition etc.
13. In this model, the algorithm must
figure out for itself
what the correct features are and
how to compute them
Before, the System Engineer
needed exhaustive information
about the domain,
in order to build a good system
14. In the rule-based system world (and even with
traditional Machine Learning) the system engineer
needed exhaustive information about the domain in
order to build a good system.
However, in this era of the IoT where new kinds of data
are becoming available at a rapid clip, Deep Learning
allows us to faster iterate on new data sources without
requiring intimate knowledge of them.
15. In the Deep Learning domain, the engineer’s main focus
is to define the architecture of the neural network. The
network needs to be large enough to have the capacity
to tune-up to a useful computation, but simple enough
so that the computation time does not exceed the
allocated time limit.
16. Optimizing the parameters of a neural network can take
days and even weeks on the strongest machines, but the
computation itself -- from raw inputs to output -- takes
a fraction of a second, and it will take exactly the same
amount of time at the end of the process as it did at the
beginning
17. Real-time intelligence and greater control agility, while at the same
time off-loading the heavy communications traffic.
18. Deep Learning is therefore a great advantage for a real-
time system like a smart sensor, because it enables
significantly enhanced scalability and flexibility. For any
given time budget, we can tailor a neural network that
fits this budget to the maximum threshold, and thus
make sure we are fully utilizing our processing power.
19. If our computational budget increases and we have more
time to run the calculation, we can assess a larger (and
presumably better) network that will fully utilize the
new budget.
20. Another advantage of using neural networks is that they
are extremely portable. Software libraries make it very
easy to build and customize a neural network, allowing
us to run the same network on different types of devices
-- just copy the parameters over and that’s it