Scribd
Upload
5 views3 pages

Lecture 4

Uploaded by

bc230414451zab
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
5 views3 pages

Lecture 4

Uploaded by

bc230414451zab
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

AI COURSE (BY IRFAN MALIK)

LECTURE 04

GOOGLE BARD
 Bard is a conversational AI chat service
developed by Google.
 While ChatGPT relies on its pre-trained
knowledge, Bard puts information from the
internet in the real time.
 Google Bard is powered Google’s very own and
most advanced large language model (LLM)
PaLM.
 https://bard.google.com/chat

WHAT IS A LANGUAGE MODEL?


A language model is an artificial intelligence
model that predicts the probability of sequence
of words or tokens in the given contest.

The dogs _____ across the park.


run Language model Probability=0.3
runs
Language model Probability=0.02
0.3>0.002
GENERATIVE AI
The type of AI which generates the content,
neither retrieve nor discriminate the content is
known as Generative AI.
There are two parts of generative model one is
ENCODER and the other is DECODER.

ENCODER
The job of encoder is to convert human centric
representation into machine centric
representation.
DECODER
The job of decoder is to convert machine
centric representation into human centric
representation.
SOME GENERATIVE MODELS
 Auto Encoders
 GAN (Generative Adversarial Net)
 LLM (Large Language Models)
 Diffusion Models

REPRESENTATION
Machine Learning always needs numerical
representation to perform task. The generative
models contain strength to learn that type of
representation which can be used to perform
many tasks.
TRANSFORMER NUERAL
NETWORKS
The most popular generative model which is in
use nowadays is Transformer Neural Networks.
It also contains encoder and decoder.

INTRODUCTION TO STABLE
DIFFUSION
 A cutting-edge image generation tool utilizing
latent diffusion models.
 Generates realistic and high-quality images
through advanced diffusion techniques.
 Allows user to explore the possibilities of image
generation by controlling diffusion parameters.
 Used in various fields including art, design and
computer vision for diverse image generation
task.

You might also like