You are on page 1of 16

J.A.R.V.I.

S
A WEBINAR BY KAUSTAV SANYAL
ASSISTANT PROFESSOR
BENGAL INSTITUTE OF SCIENCE AND
TECHNOLOGY
Just A Rather Very Intelligent
Intelligence
System
• Learn from
Experience
• Recognize Problems
• Solve Problems
General Intelligence Triarchic
01 A General Cognitive ability
03 Theory
•Analytical
that can be measured and
Numerically expressed •Creative
•Practical

Primary Multiple Intelligence


02 Abilities
•Associative memory 04 •Bodily kinesthetic

•Numerical ability •Interpersonal

•Perceptual speed •Intrapersonal

•Reasoning •Logical-Mathematical

•Spatial visualisation •Musical

• Verbal comprehension •Naturalistic

•Word Fluency •Verbal-linguistic


•Visual-spatial
Then
•Simulation of Human Intelligence in
machines.
•Machines programmed to think humanly.

What
•Machines programmed to mimic human
actions.
•Machines with ability to learn.
•Machines with ability to think rationally.
•Machines with ability to solve problems. is AI?
AI is not only about ROBOTS!
There is no unifying paradigm for AI

On the basis of training


Training the machine to Storing the training sets
and learning testing the
simulate intelligence of to simulate learning
machine of its problem
its own behavior
solving efficiency

TRAINING LEARNING SOLVING


A
THE
B
IMITATION
GAME C
NEGOTIATOR

“CAN MACHINES THINK?”

“COMPUTING MACHINERY AND


INTELLIGENCE” 1950, ALAN TURING.
DETERMINISTIC FINITE AUTOMATON
String being Processed

C $ Input Tape

Reading Head

Finite
Control

Input Tape Reading Head

Finite Control Transition


HOW DOES IT WORK?

q 0

Design of a Finite 1
Automaton with S={0,1}
accepts those strings which
starts with 1 and ends with q 1 1
0.
1 0

q 2
0
THE
MARVEL OF
PANINI
ASHTADHY
AYI
THE WELL CONSTRUCTED GRAMMAR
In total 4,000 commands are divided into 6 categories (or 7 including the “Nisedha”) to construct the most complete
possible grammar for a language. There exist no possible combinations/exceptions in the complete language which
cannot be explained otherwise. The resemblance of these commands with the idea of constructing a Computer is
remarkable.

01 02 03 04 05 06
RULES OF DOMAIN
VARIABLE METADATA RULES OF IDEA OF
IMPLEMENTAT DEFINITION
DEFINITION TRANSITION COMAPRISON
ION
THE
THE
THE SAMGYA PARIBHASHA THE VIDHI THE ATIDESHA
THE NIYAMA ADHIKARA
SUTRAS SUTRAS SUTRAS SUTRAS
SUTRAS SUTRAS
CONSTRUCTION OF THE MOTHER
SCRIPT
Every Chapter Includes all the
Consists of Four Possible Combinations
Sub Sections for a Language

1 2 3 4 5 6 7 8

VARIABLE RULES OF COMBINATION OF


DEFINITION & IMPLEMENTAT VARIABLES AND
METADATA ION IMPLEMENTATION
DEFINES THE THE VERBS POSSIBILITIES
CONSTRUCTION OF AND THER REVEALED
SENTENCES AND WORDS
CONSTRUCTS
MASTER MNEMONIC SET (SHIVA
SUTRAs)
Generation of Context- Forteen distinct linear The replacement rules for
sensitive grammar rules functions to compose the building any combination of
far before Chomsky completeLINEAR
set of phonemes TM recognizable language
CONTEXT- FUNCTIONS PRATYAHARAS
SENSITIVE RULES

Naming every mnemonic


More than one rule Well defined auxiliary accordingly to give 42
combined to build a markers to mark “start” and number of codes for the rest
unique one “end” points
AUXILIARY of the Script
NOMENCLATU
COMBINATION MARKERS RE
PANINI GRAMMAR AS A FINITE
AUTOMATON
Derived Word
As Feminine

Af
fix
of

of
fi x e
Af inin

Va
m

ch
Definition of Fe

an
the Word

&
Vib
on
f in i t i Aff

hak
D e ix o
of f Va
Affix

ti
Root cha
n &V
ibh
akt Word in
Verb i
Affi Implementation
x of (First, Second,
Verb
Third person
variations)
RECALL THE DFA

q 0

Design of a Finite 1
Automaton with S={0,1}
accepts those strings which
starts with 1 and ends with q 1 1
0.
1 0

q 2
0
WHY SANSKRIT?

NLP PRE-DEFINED
CONVINIENT
Conversion of Natural STRUCTURE
Sanskrit grammar is a very well
Language into Machine designed and developed structure
Language is efficient with which covers a vast area of
Sanskrit systematic communication

UNAMBIGUOUS VERSITILITY
The unambiguity in the The application of the
nature of this language rules are versatile in
provides better nature
opportunities for AI
THANK
YOU!
Do you have any questions?

kamanofrythm@gmail.com
+91 9800447114

CREDITS: This presentation template was created by Slidesgo,


including icons by Flaticon, and infographics & images by Freepik.

You might also like