Lecture 3
0 likes•23 views
The document discusses the limitations of perceptrons and how multilayer perceptrons can solve more complex problems. It explains that while a single perceptron can only learn linearly separable problems, multilayer perceptrons using techniques like backpropagation can learn nonlinear and more complex relationships. The document also provides information on decision boundaries and feature vectors, and how multilayer perceptrons are able to learn problems with nonlinear decision boundaries. Machine learning concepts like learning, training from data, and the growth of data are also briefly covered.
Lecture 3
- 2. Perceptron
- 3. Limitations of the Perceptron 25-Sep-17 3
- 4. Perceptron Limitations 25-Sep-17 Multi-layer perceptron can solve this problem More than one layer of perceptrons (with a hard- limiting activation function) can learn any Boolean function However, a learning algorithm for multi-layer perceptrons has not been developed until much later Back propagation algorithm (replacing the hard-limiter with a sigmoid activation function) 4
- 8. Decision Boundaries 25-Sep-17 In general, a pattern classifier carves up (partitions) the feature space into volumes called decision regions. All feature vectors in a decision region are assigned to the same category. The decision regions are often simply connected, but they can be multiply connected as well, consisting of two or more non-touching regions. In pattern recognition and machine learning, a feature vector is an n- dimensional vector of numerical features that represent some object. Many algorithms in machine learning require a numerical representation of objects, since such representations facilitate processing and statistical analysis. 8
- 10. The Multi-Layer Perceptron 25-Sep-17 A multilayer perceptron (MLP) is a feed-forward artificial neural network that generates a set of outputs from a set of inputs. An MLP is characterized by several layers of input nodes connected as a directed graph between the input and output layers. MLP uses backpropogation for training the network. MLP is a deep learning method. 10
- 11. MLP Decision Boundary – Nonlinear Problems, Solved! 25-Sep-17 11
- 12. Machine Learning 25-Sep-17 What is Learning? Learning takes place as a result of interaction between an agent and the world. The idea behind learning is that Percepts received by an agent should be used not only for acting, but also for improving the agent’s ability to behave optimally in the future to achieve the goal. 12
- 13. Why “Learn” ? 25-Sep-17 Learning is used when: Human expertise does not exist (navigating on Mars) Humans are unable to explain their expertise(speech recognition) Solution changes in time (routing on a computer network) Solution needs to be adapted to particular cases (user biometrics) 13
- 14. Things You Might Be Interested In 25-Sep-17 14
- 15. Things You Might Be Interested In 25-Sep-17 15
- 16. Learning from Data 25-Sep-17 The world is driven by data. Germany’s climate research centre generates 10 petabytes per year Google processes 24 petabytes per day The Large Hadron Collider produces 60 gigabytes per minute (~12 DVDs) There are over 50m credit card transactions a day in the US alone. 16
- 17. If Data Had mass, The Earth Would Be A Black Hole 25-Sep-17 Around the world, computers capture and store terabytes of data everyday. Science has also taken advantage of the ability of computers to store massive amount of data. The size and complexity of these data sets means that humans are unable to extract useful information from them. 17