Abstract:
Kernel methods have thrived for 20 years. Big data has greatly increased data collection. Existing kernel methods are not scalable for training or predicting. This paper introduces a general sparse kernel learning formulation based on the random feature approximation with possibly non-convex loss functions to address this challenge. We also use the orthogonal random feature approximation formulation to reduce experiment scale. We propose an asynchronous parallel doubly stochastic algorithm for large-scale sparse kernel learning (AsyDSSKL). AsyDSSKL is the first algorithm to combine asynchronous parallel computation and doubly stochastic optimization. AsyDSSKL has a full convergence guarantee. Importantly, our AsyDSSKL method outperforms existing kernel methods in training and predicting computational efficiency on large-scale real-world datasets.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
