Timezone: »

 
Spotlight
Adaptive Second Order Coresets for Data-efficient Machine Learning
Omead Pooladzandi · David Davini · Baharan Mirzasoleiman

Thu Jul 21 10:55 AM -- 11:00 AM (PDT) @ Room 318 - 320
in Optimization/Reinforcement Learning »

Training machine learning models on massive datasets incurs substantialcomputational costs. To alleviate such costs, there has been a sustained effort to develop data-efficient training methods that can carefully select subsets of the training examples that generalize on par with the full training data. However, existing methods are limited in providing theoretical guarantees for the quality of the models trained on the extracted subsets, and may perform poorly in practice. We propose AdaCore, a method that leverages the geometry of the data to extract subsets of the training examples for efficient machine learning. The key idea behind our method is to dynamically approximate the curvature of the loss function via an exponentially-averaged estimate of the Hessian to select weighted subsets (coresets) that provide a close approximation of the full gradient preconditioned with the Hessian. We prove rigorous guarantees for the convergence of various first and second-order methods applied to the subsets chosen by AdaCore. Our extensive experiments show that AdaCore extracts coresets with higher quality compared to baselines and speeds up training of convex and non-convex machine learning models, such as logistic regression and neural networks, by over 2.9x over the full data and 4.5x over random subsets.

Keywords: DL: Algorithms · DL: Everything Else · OPT: Discrete and Combinatorial Optimization · OPT: Sampling and Optimization

Author Information

Omead Pooladzandi (University of California, Los Angeles)
David Davini (UCLA)
Baharan Mirzasoleiman (Stanford University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors