I work in IBM Watson in the field of Cognitive Computing. It is based on various techniques in Artificial Intelligence including Machine Learning, Natural Language processing, Speech processing, Predictive and Content Analytics among other things. There is a distinction though. For example in my paper on the use of Watson for Personal Wealth Management you will find that Watson and Cognitive Computing the system provides the analyst with information and confidence based next best actions instead of just telling the analyst what to do or how to invest the client's money. Here the system is an assistant to the human and the human is always the final decision maker even if the system is almost always right.
On the left please click on the various fields or the products to know more about what I do in each area... for example Watson Content Analytics or WCA, Watson Discovery Advisor or WDA, Natural Language Processing etc.
Every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it. An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans and improve themselves. - John McCarthy. (1955) [1]
A machine or computer program that learns how to do tasks that require forms of intelligence and are usually done by humans. [1]
The original 7 aspects of A.I [1]
1. Simulating higher functions of the human brain
2. Programming a computer to use general language
3. Arranging hypothetical neurons in a manner so that they can form concepts.
4. A way to determine and measure problem complexity
5. Self improvement
6. Abstraction: Defined as the quality of dealing with ideas rather than events.
7. Randomness and creativity
Intelligence a definition by Jack Copeland [1]
Generalized Learning
Reasoning
Problem Solving
Perception
Language Understanding
Some examples of AI [1]
-Machine Learning
-Computer Vision
-NLP
-Robotics
-Pattern Recognition
-Knowledge Management
Artificial Intelligence is a Science and Machine Learning is a subset a practical implementation of Artificial Intelligence.
Strong AI and Weak AI [1]
Strong AI: Simulating the human brain and build systems that think and give us an insight into how the brain works
Weak AI: A system that behaves like a human but doesn’t give an insight into how the brain works
IBM Watson: In between Strong and Weak AI it is a system that is inspired by human reasoning but doesn’t have to stick to it. For example IBM Watson. Reads Info, recognizes patterns and builds evidence and provides answers with a confidence level measure. [1]
Google Deep learning: Uses Neural Networks.
Neural Networks are a subset of machine learning
Expert System: System that employs human knowledge in a computer to solve problems that ordinarily require human expertise. Practical Application of A knowledge database. Eg: DeepMinds’ AlphaGo. [1]
[1]
Narrow or simple AI systems (Deep Blue, SIRI, FB Friend recommendations) [2]
For learning we need data and we need good models for learning.
Neural Networks is a type of Machine Learning. It had layers and nodes. And learning is the fine tuning of these to arrive at a conclusion. When the information propagates from left to right its called Forward Propagation Network. When error rates etc are sent backwards for further learning its called Back Propagation Network.
Images are very well suited for neural networks. When information extracted is a part of the original image as a layer such that it can be blended back into the original it is a convolutional neural network. There are multiple layers of these perceptrons in a convolutional neural network. CNNs have multi-layers of receptive fieldsø each is a small neuron collectionø each collection has an output that is a tileø all these are tiled to form a high res image of the original.
A multilayer perceptron (MLP) is a feedforward artificial neural network model that maps sets of input data onto a set of appropriate outputs. An MLP consists of multiple layers of nodes in a directed graph, with each layer fully connected to the next one. Except for the input nodes, each node is a neuron (or processing element) with a nonlinear activation function. MLP utilizes a supervised learning technique called backpropagation for training the network
Convolutional neural networks are a part of Deep Learning which is a part of Neural Networks which is a part of Machine Learning which is a practical implementation of Artificial Intelligence.
Copyright (c) 2016 vijaigandikota.com. All rights reserved. | Photos by Vijai Gandikota | Design by Vijai Gandikota.com.
