Computational Learning An intuitive approach Human Learning Objects

Computational Learning An intuitive approach Human Learning  Objects

Computational Learning An intuitive approach Human Learning Objects in world Learning by exploration and who knows? Language

informal training, inputs may be incorrect Programming A couple of examples of loops or recursions Medicine See one, do one, teach one

People: few complex examples, informal, complex behavioral output Computational Learning Representation provided Simple inputs: vectors of values Simple outputs: e.g. yes or no, a number, a disease

Many examples (thousands to millions) Quantifiable + Useful, e.g. automatic generation of expert systems Concerns Generalization accuracy Performance on unseen data

Evaluation Noise and overfitting Biases of representation You only find what you look for. Three Learning Problems Classification: from known examples

create decision procedure to guess class Patient data -> guess disease Regression: from known examples create decision procedure to guess real numbers Stock data -> guess price Clustering: putting data into meaningful

groups Patient Data -> new diseases Simple data attribute-value representation = 1 example

Sex, age, smoker, etc are the attributes Values are male, 50, true etc Only data of this form allowed. The Data: squares and circles ? ?

? ? Linear Decision Boundaries Perceptron Learning a (hyper)-line Given data

Construct line the decision boundary Usually defined by a normal n data is on one side if dot product data * n >0 Recall * is x1*y1+x2*y2. What a neuron does SVM: Optimization approach

middle line Decision Trees: rectangular boundaries 1-Nearest Neighbor Piecewise Linear boundaries

1-Nearest Neighbor classification If x is a example, find the nearest neighbor NN in the data using euclidean distance. Guess the class of c is the class of NN K-nearest neighbor: let the k-nearest neighbors vote Renamed as IB-k in Weka

Neural Nets Smooth boundaries Neural Net A single perceptron cant learn some simple concepts, like XOR A multilayered network of perceptrons can learn any boolean function

Learning is not biological but follows from multivariable calculus Gedanken experiments

Try ML algorithms on imagined data Ex. Concept: x>y, ie. Data looks like 3,1,+. 2,4,-. etc Which algorithms do best? And how well?

Consider the boundaries. My guesses: SMO> Perceptron>NearestN>DT. Check Guesses with Weka

199 examples. DT= 92.9 (called J48 in weka) NN= 97.5 (called IB1 in weka) SVM = 99.0 (called SMO in weka)

Recently Viewed Presentations

  • Hemoglobin

    Hemoglobin

    Hemoglobin is a globular hemoprotein. Hemeproteinsare a group of specialized proteins that contain hemeas a tightly bound prosthetic group.. Heme. is a complex of . protoporphyrin. IX & ferrous . iron (Fe2+) . The . iron. is held in the...
  • Activated Sludge Modeling - Iowa State University

    Activated Sludge Modeling - Iowa State University

    Particulate removal rate. ... Monod Equation and Unified Model. One of the main things to realize about the traditional model is that it only accounts for the biochemical reactions of readily degradable soluble substrate at steady state conditions. It's a...
  • Differentiated Event Help pack - WJEC

    Differentiated Event Help pack - WJEC

    WJEC. GCSE Hospitality and Catering . Unit 3. Student information . UNIT 3 EVENT . You will be split into mixed ability groups to complete this task. It is not possible for you all to do the task together as...
  • Introduction to the Finite Element Method

    Introduction to the Finite Element Method

    Times New Roman Default Design The Finite Element Method Course Outline Course Outline con't Course outline con't Course outline con't Course outline con't Course outline con't Course outline con't Course outline con't
  • ImmPact - maine.gov

    ImmPact - maine.gov

    Dose Decrementing - What? Why? What? A comprehensive vaccine management system (dose decrementing) that links the ImmPact application functions to vaccine ordering, inventory management, and vaccine administration with the ability to auto dose decrement vaccines from inventory as vaccines are...
  • River Landforms

    River Landforms

    hogback escarpment. is a long sharp ridge of a steeply dipping layer (or layers) of harder igneous or sedimentary rock. Surrounding, weaker rock has been eroded away, leaving a steep ridge of more resistant material.
  • Depressants - Riverside City College

    Depressants - Riverside City College

    Depressants . A depressant, or central depressant, is a drug that lowers neurotransmission levels, which is to depress or reduce arousal or stimulation, in various areas of the brain. Depressants are also occasionally referred to as "downers" as they lower...
  • 1.3 1. 2. 3. 4. Adding and Subtracting

    1.3 1. 2. 3. 4. Adding and Subtracting

    Keep the same denominator Simplify. Adding Fractions To add fractions with different denominators: 1. Write each fraction as an equivalent fraction with the LCD. 2. Add the numerators and keep the LCD. 3. Simplify.