Presentation is loading. Please wait.

Presentation is loading. Please wait.

Machine Learning Case study. What is ML ?  The goal of machine learning is to build computer systems that can adapt and learn from their experience.”

Similar presentations


Presentation on theme: "Machine Learning Case study. What is ML ?  The goal of machine learning is to build computer systems that can adapt and learn from their experience.”"— Presentation transcript:

1 Machine Learning Case study

2 What is ML ?  The goal of machine learning is to build computer systems that can adapt and learn from their experience.”  When a computer system improve its performance at a given task overtime, without re-programming, it can be said to have learned something.

3 What is ML ? Learning, like intelligence, covers such abroad range of processor that it is difficult to define precisely. AI: Learning as a search problem, using prior knowledge to guide learning. ML :is the key to machine intelligence just as human learning is the key to human intelligence.

4 ML refers to changes in systems that perform tasks associated with AI. prediction diagnosis planningrobot control recognition ML might help us understand how animals and human learn

5 Why Machine Learning? 1.(Relatively) new kind of capability for computers 1.Data mining: extracting new information from medical records, maintenance records, etc. 2.Self-customizing programs: Web browser that learns what you like and seeks it out

6 Why Machine Learning?(cont) 3. The amount of knowledge available about certain tasks might be too large for explicit encoding by human. 4.Applications we can’t program by hand: E.g. speech recognition, autonomous driving

7 Why Machine Learning?(cont) 5. Environments can change over time. Machine that can adapt to a changing environment would reduce the need for constant redesign.

8 Application of ML Content-Based Image Retrieval Given database of hundreds of thousands of images How can users easily find what they want? One idea: Users query database by image content –E.g. “give me images with a waterfall”

9 Content-Based Image Retrieval (cont’d) Better approach: Query by example –Users give examples of images they want –Program determines what’s common among them and finds more like them

10 Content-Based Image Retrieval (cont’d) User’s Query: System’s Response: Yes NO!User Feedback:

11 Content-Based Image Retrieval (cont’d) User’s feedback then labels the new images, which are used as more training examples, yielding a new hypothesis, and more images are retrieved

12 How Does the System Work? For each pixel in the image, extract its color + the colors of its neighbors These colors (and their relative positions in the image) are the features the learner uses (replacing e.g. number-of-wheels) A learning algorithm takes examples of what the user wants, produces a hypothesis of what’s common among them, and uses it to label new images

13 Problems of ML 1) Getting it to work. 2) It must be have some kind of success of certain kind. 3) Computer produced knowledge are not always certain.

14 Thanks for listening


Download ppt "Machine Learning Case study. What is ML ?  The goal of machine learning is to build computer systems that can adapt and learn from their experience.”"

Similar presentations


Ads by Google