Ch 6 - Deep Learning

Application of Deep Learning

Customer Support, Medical Care, Self-Driving Cars

Customer Service with AI

The system continuously learns and over time predicts the likely answer. It will then soon be able to offer solutions directly to the customer.

Machine Language

a subset of AI and a technique to achieve AI through algorithms trained with data

Deep Learning is

a subset of machine learning which is a subset of artificial intelligence

Cognitive Computing

combines imagination and emotion with machine (pattern recognition, mass information)

Deep Learning Process

features are picked out by the ANN with no human intervention, requires high volume of data

Examples of deep learning

google translate to another language, cellphone gallery grouping pictures based on location

Machine Language Process requires

human intervention to produce an output, dataset can be small

Strong AI

learns like a human and can solve human problems (deep learning)

Deep Learning

mimicking what the brain does. It is an artificial Neural Network

Cognition

process of acquiring knowledge through thoughts, experiences and senses

Limitations of Deep Learning

requirement of large data, computational power requiring a GPU, training time takes hours or months

Micro routing

routing directly to the best agent, no guess work

Artificial Intelligence (AI)

technique for a machine to mimic human behavior

The activation functions will allow

the deep learning process to determine the output through analysis of layers.

Customer Service with AI shortens

the time between ticket receival and agent assignment to get to solutions faster.

Biases are activated where

they are highlighted which then passes on the information to the next layer.

Steps of a Neural Network

weighted sum of inputs is calculated, bias is added, result is fed to an activation function, specific neuron is activated

Weak AI

will only do what they are programmed to do (machine learning)