1 / 7

IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING

IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING. Lazise, Garda Lake, Italy, 24-28 September 2007. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCIENCE ON LEARNING. GROUP 9. LUIS PAYA LUCA LONINI ALIREZA DERAKHSHAN. WHAT WE HAVE LEARNED. NOVELTY DETECTION.

hester
Download Presentation

IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING Lazise, Garda Lake, Italy, 24-28 September 2007 IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCIENCE ON LEARNING GROUP 9 LUIS PAYA LUCA LONINI ALIREZA DERAKHSHAN • WHAT WE HAVE LEARNED. • NOVELTY DETECTION. • LEARNING WITH RECURRENT NEURAL NETWORK WITH PARAMETRIC BIASES. • APPLICATIONS IN OUR RESEARCH AREA

  2. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING 1. WHAT WE HAVE LEARNED Lazise, Garda Lake, Italy, 24-28 September 2007 • Use of a simple robotic platform to carry out experiments in complex techniques of machine learning. • We have dealt with simple external information - more complex information should be added e.g. more sensory data. • Learning by imitation • Analytical models (system identification, policy learning by imitation). • Non Analytical models (learning with recurrent neural networks with parametric biases). • Statistical Analysis and Data Mining with Orange.

  3. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING 2. NOVELTY DET ECTION Lazise, Garda Lake, Italy, 24-28 September 2007 • Working with readings from a Magellan’s 16 sonar sensors in a wall following behavior. • 1st train: • s = 12 · (standard deviation) • q = 0.6 • 58 kernels in the model base. • Distances of each test data to the nearest kernel of the model base. • Not a clear novelty among the test data.

  4. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING 2. NOVELTY DETECTION Lazise, Garda Lake, Italy, 24-28 September 2007 • 2nd train: • s = 7 · (standard deviation) • q = 0.6 • 321 kernels in the model base. • Distances of each test data to the nearest kernel of the model base. • Two possible candidates. • The 2nd one (reading 100) is the novelty one (maximum distance to the nearest kernel).

  5. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING 3. LEARNING WITH RECURRENT NEURAL NETWORK WITH PARAMETRIC BIASES. Lazise, Garda Lake, Italy, 24-28 September 2007 • Building complex behaviors by combining simple primitive behaviors. • Each simple primitive can be coded with 2 biases. Biases: [0.68 0.40] Sinusoid [0.73 0.36] Left [0.19 0.78] Right Biases: Keep Object Left [0.99 0.0]

  6. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING 3. LEARNING WITH RECURRENT NEURAL NETWORK WITH PARAMETRIC BIASES. Lazise, Garda Lake, Italy, 24-28 September 2007 • Adding new primitives is possible Biases: Obstacle Avoidance [0.08 0.29]

  7. IEEE-RAS / IFRR SCHOOL OF ROBOTICS SCINECE ON LEARNING Applications in OUR Research Area Lazise, Garda Lake, Italy, 24-28 September 2007 • Appearance-based Navigation • These techniques can be applied to the localization and navigation of a mobile robot using more complex information (e.g. The information of the whole scene, laser measures, etc.). • It is necessary to analyze the scene and extract the most relevant information. • Classification of Playing Behavior • Novelty Detection can be applied to categorize different Playing Behavior based on some reference behaviors. • Human motor learning models • Machine learning techniques and Experiments with robots can be useful to test hypothesis on neuroscientific theories on how we do organize movements • Novel control techniques can be applied to new generation robots

More Related