Some Properties of the Gaussian Kernel for One Class Learning

“…l is the number of observations of the training features x k , and Φ(x k ) is a mapping from the original feature values to a higher dimensional space F. Values of ξ k are slack variables in the optimization problem, and ν represents the fraction of examples allowed to be considered outliers of the volume. We apply a Gaussian kernel and tune the kernel parameter σ using a heuristic based on the width of the training data set, as described by [10]. Rabaoui et al [25] proposed a method for combining multiple OCSVM classifiers into a single classifier capable of multi-category classification.…”

Section: Ocsvm Trainingmentioning

confidence: 99%

Methods for robotic tool-mediated haptic surface recognition

Romano

Kuchenbecker

2014

2014 IEEE Haptics Symposium (HAPTICS)

View full text Add to dashboard Cite

Robots need to be able understand the haptic sensations that occur during physical interaction in order to recognize and manipulate objects. This paper centers on the problem of robotic toolmediated haptic surface recognition. We collected data as a PR2 robot interacted with fifteen different surfaces through a novel tool that captures high-definition recordings of tactile vibrations, contact forces, and tool position. We then developed algorithms for using the recorded data to recognize the identity of the surface under varying contact conditions. We show that successful recognition of surfaces touched through a tool critically depends on accounting for the physical contact information (tool speed and normal force). Additionally, we present several enhancements to improve surface recognition including the combination of multiple vibration axes into a single axis, using frictional force during contact, and logarithmically spaced frequency scaling for vibration analysis. We implement a flexible classifier composed of multiple One-Class Support Vector Machines trained on a data set containing all fifteen surfaces, and we then use this classifier to identify these same surfaces in a new data set recorded under substantially different contact conditions, achieving an overall texture recognition rate of 80%.

show abstract

“…The Gaussian function with mean dsf i and variance 0.04 is defined as the kernel K j in Equation (7). The Gaussian kernel is a powerful kernel widely used in pattern recognition [54]. The variance is chosen to be the smallest value that makes the empirical fragility functions increase monotonically.…”

Section: Development Of Fragility Functions For Different Damage Statesmentioning

confidence: 99%

Development of fragility functions as a damage classification/prediction method for steel moment‐resisting frames using a wavelet‐based damage sensitive feature

Noh

Lignos

Nair³

et al. 2011

Earthq Engng Struct Dyn

View full text Add to dashboard Cite

SUMMARYFragility functions are commonly used in performance-based earthquake engineering for predicting the damage state of a structure subjected to an earthquake. This process often involves estimating the structural damage as a function of structural response, such as the story drift ratio and the peak floor absolute acceleration. In this paper, a new framework is proposed to develop fragility functions to be used as a damage classification/prediction method for steel structures based on a wavelet-based damage sensitive feature (DSF). DSFs are often used in structural health monitoring as an indicator of the damage state of the structure, and they are easily estimated from recorded structural responses. The proposed framework for damage classification of steel structures subjected to earthquakes is demonstrated and validated with a set of numerically simulated data for a four-story steel moment-resisting frame designed based on current seismic provisions. It is shown that the damage state of the frame is predicted with less variance using the fragility functions derived from the wavelet-based DSF than it is with fragility functions derived from an alternate acceleration-based measure, the spectral acceleration at the first mode period of the structure. Therefore, the fragility functions derived from the wavelet-based DSF can be used as a probabilistic damage classification model in the field of structural health monitoring and an alternative damage prediction model in the field of performance-based earthquake engineering. Copyright

show abstract

Some Properties of the Gaussian Kernel for One Class Learning

Cited by 55 publications

References 11 publications

Support Vector Machine for Network Intrusion and Cyber-Attack Detection

Support Vector Machine for Network Intrusion and Cyber-Attack Detection

Methods for robotic tool-mediated haptic surface recognition

Development of fragility functions as a damage classification/prediction method for steel moment‐resisting frames using a wavelet‐based damage sensitive feature

Contact Info

Product

Resources

About