Scribd
Upload
42 views

00-Introduction DNN

Uploaded by

kamble793supriya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
42 views

00-Introduction DNN

Uploaded by

kamble793supriya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 32

CMPE 258

Deep Learning
Prerequisites
• CMPE 255 or CMPE 257 or instructor consent. Computer Engineering
and Software Engineering majors only.

• Familiarity with Python


• Strong organizational and team skills
Machine Learning
• Field of study that gives computers the ability to learn
without being explicitly programmed
By Arthur Samuel (1959)
What is deep learning?
• Deep Learning vs Machine Learning
• What if small data set?
• Why is it powerful? Is it difficult?
• What is transfer learning?
• Is CNN deep learning?
• https://en.wikipedia.org/wiki/Deep_learning
IBM Deep Blue vs. Kasparov (1996 & 1997)
IBM Watson in Jeopardy! (2011)
Fast R-CNN Demo
Mobileye
AlphaGo vs. Lee, Sedol (2016)
Language Understanding
ChatGPT https://openai.com/blog/chatgpt/
Machine Learning
Neural Network
• Collection of connected units (neurons), inspired by
the biological neural networks that constitute human
brains
Neural Networks?
History
• 1943: McCulloch–Pitts “neuron”
• Started the field
• 1962: Rosenblatt’s perceptron
• Learned its own weight values; convergence proof
• 1969: Minsky & Papert book on perceptrons
• Proved limitations of single-layer perceptron networks
• 1982: Hopfield and convergence in symmetric networks
• Introduced energy-function concept
• 1986: Backpropagation of errors
• Method for training multilayer networks
• Present: Probabilistic interpretations, Bayesian and spiking networks
The network
between
biological
neurons
(From: Wikipedia)
Biological inspiration
Dendrites

Soma (cell body)

Axon
Source: Peter Andras
Artificial neurons
The McCullogh-Pitts model:
• spikes are interpreted as spike rates;
• synaptic strength are translated as synaptic weights;
• excitation means positive product between the incoming spike
rate and the corresponding synaptic weight;
• inhibition means negative product between the incoming spike
rate and the corresponding synaptic weight;

Source: Peter Andras


Biological inspiration

dendrites
axon

synapses

The information transmission happens at the synapses.

Source: Peter Andras


Learning in biological systems
Learning = learning by adaptation
The young animal learns that the green fruits are sour, while the
yellowish/reddish ones are sweet. The learning happens by
adapting the fruit picking behavior.
At the neural level the learning happens by changing of the
synaptic strengths, eliminating some synapses, and building new
ones.
Source: Peter Andras
“Deep” Neural Network
Artificial neurons
Neurons work by processing information. They receive and
provide information in form of spikes.
x1
x2
w1
x3
n
Output
… w2
z =  wi xi ; y = H ( z )
Inputs

xn-1 y
w3 i =1

xn .
.
. wn-1

wn

The McCullogh-Pitts model Source: Peter Andras


Backpropagation for training
Geoffrey Hinton
• Pre-training for deep neural networks
Deep Learning for AI
• July 2021 CACM: Deep Learning for AI
• https://www.youtube.com/watch?v=at0OXoUmEj4
ML, NN & DL Summary
• Machine Learning
• Field of study that gives computers the ability to learn without being explicitly
programmed
• Neural network
• One of ML classifiers, being trained by backpropagation
• Extension of logistic & softmax regression
• Deep Learning
• Framework of using neural networks with deep (many) layers
• Most of state-of-the-art ML-based products utilized DL!
Image Classification

Copyright S. Shim 30
Deep
Reinforcement
Learning
• Google DeepMind
‘Agent 57’ Beats
Human Baselines
Across Atari Games
Suite
Autonomous Robots

Any Questions?
Copyright S. Shim 32

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy