310CT Intelligent Agents Assignment 1

INTI International College Penang School of Engineering and Technology

3+0 Bachelor of Science (Hons) in Computer Science, in collaboration with Coventry University, UK
3+0 Bachelor of Science (Hons) in Computing, in collaboration with Coventry University, UK

Coursework cover sheet

Section A - To be completed by the student

Full Name:

CU Student ID Number:

Semester:

Lecturer: VASUKY MOHANAN

Module Code and Title: 310CT INTELLIGENT AGENTS

Assignment No. / Title: ASSIGNMENT 1 20% of Module Mark

Hand out date: 12/4/21 Due date: 10/5/21

Penalties: No late work will be accepted. If you are unable to submit coursework on time due to
extenuating circumstances you may be eligible for an extension. Please consult the lecturer.
Declaration: I/we the undersigned confirm that I/we have read and agree to abide by the
University regulations on plagiarism and cheating and Faculty coursework policies and
procedures. I/we confirm that this piece of work is my/our own. I/we consent to appropriate
storage of our work for plagiarism checking.

Signature(s): --------------------------------------

1
 310CT Intelligent Agents Assignment 1

Section B - To be completed by the module leader

Intended learning outcomes assessed by this work:

1. Critically analyze the architecture of agent-based applications.
2. Identify and critically discuss specific issues in AI research literature.
Marking scheme Max Mark

1. Group Assignment Report 100

Total 100

Lecturer’s Feedback

Internal Moderator’s Feedback

2
310CT Intelligent Agents Assignment 1

310CT Intelligent Agents Assignment 1

Group Assignment

Agent technology in Intent Based Networking

In this assignment, you are required to write a research paper into agent technology in Intent
Based Networking. Intent Based Networking, (IBN) is an emerging technology concept that
aims to apply a deeper level of intelligence and intended state insights to networking. Ideally,
these insights replace the manual processes of configuring networks and reacting to network
issues. Simply put, administrators can send a request to tell the network what outcome they
want (their intent) instead of needing to code and execute individual tasks manually.
Intent Based Networking systems not only automate time-consuming tasks and provide real-
time visibility into a network’s activity to validate a given intent, but they also predict
potential deviations to that intent, and prescribe the action required to ensure that intent. This
greater intelligence makes the network faster and more agile and reduces errors. This ability
to self-monitor and self-correct is a core component of intent-based networking.

With IBN, system administrators input their intent, and the network puts their intent into
action through the following cyclical process:
 Translation: The network translates the system administrator’s intent into network
and security policies. For example, employees in an accounting department might
need secure access to payroll information, while other employees should have limited
access to such data. With IBN, the system administrator could define these parameters,
and the network would translate them into policies. IBN’s ability to create policies
based on user or group allows system administrators to dictate complex, nuanced
policies.
 Activation: Next, the network implements those policies across the network using
automation.
 Assurance: After implementing the policies, the network monitors itself to ensure
compliance. When issues arise, the network recognizes them and intelligently
determines the best course of action to resolve them. If the network can’t reach a
resolution, it notifies systems administrators of the issue and provides
recommendations for remediation.
AI comes into play at every step in this process. First, the network uses AI to translate the
system administrator’s intent into policies and implement those policies across the network on
its own. Then, the network draws on AI to continuously monitor network activity and
intelligently analyze that data to identify problems, decide on solutions and implement them.
These steps form a loop so that the network continuously monitors itself for performance,
security or compliance issues, and system administrators can make changes at any time.
Through AI, IBN accelerates network changes, improves agility and enhances network
performance and security.

The content of the report should include the topics such as agent technology characteristics,
Intent Based Networking requirements, application of agents to Intent Based Networking
applications, and a small case study.

Assume that this research paper’s audience and purpose is to provide information to other
researchers.

3
310CT Intelligent Agents Assignment 1

Marking Scheme

Agent technology characteristics (20%)

- Characteristics of agent technology by relating to Intent Based Networking
applications. Try to include characteristics that are specific to Intent Based
Networking systems.

Intent Based Networking Systems Literature Review (20%)

- A discussion of the current state of research in Intent Based Networking systems.
Based on the discussion, including a critical analysis on the future of Intent Based
Networking.

Application of Agents to Intent Based Networking Systems (20%)

- Discussion of how agents have been implemented in Intent Based Networking
systems. This discussion must critically analyse the implementations.

Small case study, writing style and presentation of the report (40%)
- 10 % for presentation - individual
- 20% for case study-choose a particular Intent Based Networking system and discuss
role of agents type of agent architectures used, knowledge representation language
used, inference techniques used etc.
- 10% - writing style and references

Guidelines

The followings are the guidelines for obtaining marks over 70%. If you do not conform
to the following guidelines, you should expect to gain less than 40%.

Marks <40% Marks > 70%

Only <=TWO identified Identification of at least FIVE key
characteristics of agent technology.

Only <=TWO identified Specification of at least FIVE

requirements of Smart Intent Based
Networking Systems.

Lack of depth and critical analysis Critical analysis of the advantages and
drawbacks of applying agent technology
in Intent Based Networking Systems.
There must be at least 3 different
implementations discussed.

Case study is inappropriate and does not fit Proposal of an appropriate case study
the criteria of Smart Intent Based including evidence that demonstrates
Networking Systems the role of agents in Intent Based
4
310CT Intelligent Agents Assignment 1

Networking Systems
Formatting, grammatical issues Writing of the report in good academic
style

Old or insufficient reference Structure of the report supported by at

least FIVE relevant references to key
papers/sources not older than 3 years