(3 hours of instruction!)

\n\nSpeaker: C L Kim

\nOverview: From the book introduction: “Neural networks and deep learning currently provides the best solutions to many problems in im age recognition\, speech recognition\, and natural language processing.”\n

This Part 1 and the planned Part 2 (late spring/early summer 2021\, to be confirmed) series of courses will teach many of the core concepts b ehind neural networks and deep learning.

\nReference book: “Neural N
etworks and Deep Learning” by Michael Nielsen\, http://

More from the book introduction: “We’ll learn the core principles be
hind neural networks and deep learning by attacking a concrete problem: th
e problem of teaching a computer to recognize handwritten digits. …it can
be solved pretty well using a simple neural network\, with just a few tens
of lines of code\, and no special libraries.”

\n“But you don’t need
to be a professional programmer.”

\nThe code provided is in Python\,
which even if you don’t program in Python\, should be easy to understand w
ith just a little effort.

Benefits of attending the series:

\n< p>* Learn the core principles behind neural networks and deep learning.\n* See a simple python program that solves a concrete problem: teachin g a computer to recognize a handwritten digit.

\n* Improve the result through incorporating more and more of core ideas about neural networks a nd deep learning.

\n* Principle-oriented\, with worked-out proofs of fundamental equations of backpropagation for those interested.

\n* Ye t hands-on practical\, with simple code examples.\n

Course Backgroun d and Content: This is a live instructor-led introductory course on Neura l Networks and Deep Learning. It is planned to be a two-part series of cou rses. The first course is complete by itself. It will be a pre-requisite f or the planned second course. The class material is mostly from the highly -regarded and free online book “Neural Networks and Deep Learning” by Mich ael Nielsen\, plus additional material such as some proofs of fundamental equations not provided in the book\, and (in planned Part 2) touching on m ore recent neural network types such as ResNet.

\nAgenda:

\nIn troduction to Practical Neural Networks and Deep Learning (Part 1)

\nFeedforward Neural Networks.

\n* Simple (Python) Network to classif
y a handwritten digit

\n* Learning with Gradient Descent

\n* How
the backpropagation algorithm works

\n* Improving the way neural net
works learn:

\n** Cross-entropy cost function

\n** Softmax activ
ation function and log-likelihood cost function

\n** Rectified Linear
Unit

\n** Overfitting and Regularization:

\n*** L2 regularizati
on

\n*** Dropout

\n*** Artificially expanding data set

\n**
* Hyper-parameters

Introduction to Practical Neural Networks and D eep Learning (planned Part 2\, to be confirmed)

\nConvolutional Neur al Networks.

\n* Local receptive field\, Feature map. * Pooling laye r. * Simple (Python) Convolutional Neural Network to classify a handwritte n digit. * Improving the network\, Regularization. * Touch on more recent progress in image recognition\, such as Residual Network (ResNet).

\nPre-requisites: There is some heavier mathematics in proving the four fu ndamental equations behind backprogation\, so a basic familiarity with mul tivariable calculus and linear algebra is expected\, but nothing advanced is required. (The backpropagation equations can be also just accepted with out bothering with the proofs since the provided python code for the simpl e network just makes use of the equations.)

\nSpeaker Background: C L Kim works in Software Engineering at CarGurus\, Inc. He has graduate deg rees in Business Administration and in Computer and Information Science fr om the University of Pennsylvania. He has previously taught for a few year s the well-rated IEEE Boston Section class on introduction to the Android platform and API.

\nDecision (Run/Cancel) Date for this Course is
Monday\, March 15

\n\n\n\nIEEE Members $110

\nNon-members $130

DTSTART;TZID=America/New_York:20210320T090000
DTEND;TZID=America/New_York:20210320T121500
LOCATION:A live\, interactive webinar
SEQUENCE:0
SUMMARY:Introduction to Practical Neural Networks and Deep Learning (Part I
)
URL:http://ieeeboston.org/event/neuralnetworks/
X-COST-TYPE:free
END:VEVENT
END:VCALENDAR