Вы находитесь на странице: 1из 2

EE I6530 Artificial Intelligence Solutions in Electrical Engineering

Electrical Engineering Department


The City College of New York, CUNY
Course Name: EE I6530 - Artificial Intelligence Solutions in Electrical Engineering
Pre-requisites: I0100 (knowledge of linear systems and basic programming skills)
(with permission of the instructor, EE, CS, and CpE senior students can take this class as a lecture
elective)
Text: Class notes prepared by the instructor and recently published papers in the field.
Grading: Projects 30%, Midterm 30%, Final Exam 40%.
Basic rules for remotely attending this class (starting Fall 2020):

1. You have to have a reliable Internet connection to be able to take this class remotely.
2. Only CCNY email address are allowed to attend the class. Other external emails (gmail, etc.)
will be able to login into the server.
3. You will be attending classes using zoom visual. Make sure that you have a quiet room that
you can use while in remote classes.
4. During exams, if you lose connectivity to server, you will not be allowed back into the exam.

Course description:

Class Scope
Introduction to AI Principles: The fundamental concepts in artificial intelligence (AI) are
Chapter 1:
presented. A historical perspective and popular applications of AI are briefly discussed.
Genetic Algorithms: An introduction to the principles of AI is provided. Principles of
Chapter 2:
genetic algorithms (GAs) are presented. Reader is introduced to the concepts of
optimization, fitness and reproduction within the context of GAs.
Genetic Algorithms for Knapsack Class of Problems: Optimization technique known as
Chapter 3:
knapsack problem (KP) is introduced. A simple algorithm to implement KP is presented.
As a case study, a GA solving KP for a dietary menu selection application is studied. This
example uses Matlab to realize GA.
Genetic Algorithms for Controlling Autonomous Swarms: GA applications to flight
Chapter 4:
control of autonomous unmanned vehicles (UAVs) is discussed. UAV swarms for
different applications are introduced. A simple algorithm for UAV flight control using a
GA is analyzed. As a case study a GA-based flight control algorithm is presented in
Python. Impact of GA parameters on UAV performance is observed.
Genetic Algorithms for Traveling Salesman Class of Problems: A class of optimization
Chapter 5:
methods, called traveling salesman problem (TSP), together with its variations are
presented. Dependency of GA parameters on flight control precision are discussed over
a GA implementation for TSP written in Python.
Artificial Neural Networks: This chapter presents ANNs, among the most popular AI
Chapter 6:
tools for recent applications, their main components and their working principles. Types
of ANNs and examples of application to different problems are discussed.
Artificial Neural Networks for Outcome Prediction:
Chapter 7:
Outcome prediction based on historical data is one of the most common applications
for ANNs. This chapter studies two such examples: prediction of exam grade and
prediction of bike usage. Both problems are implemented using Python.
Artificial Neural Networks for Algorithmic Trading:
Chapter 8:
ANNs are widely used in finance industry especially as essential tools for algorithmic
trading. In this chapter a few technical indicators defined for financial stock data are
introduced. A simple algorithmic trading policy is implemented in Python.
Artificial Neural Networks for Natural Language Processing: NLP involves methods for
Chapter 9:
identification of a relationship between spoken words or written text and external
events. We show how to employ ANNs for NLP. In this chapter we investigate possible
relationship between daily announcements from a government and stock market
movements. ANN implementation in Python is used for this study.
Convolutional Neural Networks: This chapter introduces CNNs, together with their
Chapter
structure and functionality of components.
10:
Convolutional Neural Networks for Optical Character Recognition: OCR is a widely
Chapter
used automation tool for many daily applications. This chapter presents a OCR
11:
implementation using CNNs, which recognized hand drawn characters in any language
that the CNN is trained for. The case study is coded in Matlab.
Convolutional Neural Networks for Speech Recognition: CNN can be an effective tool
Chapter
for realizing speech recognition software. This chapter presents a speech recognition
12:
system that can translate words to different languages. Example is programmed in
Matlab.
Chapter Particle Swarm Optimization: PSO is a multi-objective optimization method
that is capable of solving complex problems with multiple goals. In this chapter, a
13:
simple PSO algorithm is presented. PSO reproduction rules are explained. Power
of PSO is demonstrated by solving a multi-dimensional mathematical benchmark
problem.
Chapter Differential Evolution: DE is designed to be a computationally light AI method
by changing the traditional reproduction rules in GA. It is a useful alternative to
14:
the other evolutionary algorithms especially when generating results quickly is
crucial.
Chapter Grammatical Evolution: GE provides candidate solutions as syntactically
correct programs that can be implemented in any programming language. It is
15:
powerful technic for cases where many different parameters are to be considered
to obtain optimum solutions.
Chapter AI in Computational Biology: AI based techniques can be applied to many tasks
in computational biology. Prediction of prognosis for personalized care of
16:
progressive illnesses such as cancer and Alzheimer’s disease, prescription of
personalized drug regimen and drug efficacy studies are among such tasks.
Patient gene expressions, blood tests, tumor size measurements, brain scans and
MRI are among the types of inputs used by AI methods. Chapter presents results
of a case study for breast cancer patients obtained by a GA-based approach.

Вам также может понравиться