Timezone: »

 
Poster
Provable Data Subset Selection For Efficient Neural Networks Training
Morad Tukan · Samson Zhou · Alaa Maalouf · Daniela Rus · Vladimir Braverman · Dan Feldman

Thu Jul 27 01:30 PM -- 03:00 PM (PDT) @ Exhibit Hall 1 #813
in Poster Session 5 »

Radial basis function neural networks (RBFNN) are well-known for their capability to approximate any continuous function on a closed bounded set with arbitrary precision given enough hidden neurons. In this paper, we introduce the first algorithm to construct coresets for RBFNNs, i.e., small weighted subsets that approximate the loss of the input data on any radial basis function network and thus approximate any function defined by an RBFNN on the larger input data. In particular, we construct coresets for radial basis and Laplacian loss functions. We then use our coresets to obtain a provable data subset selection algorithm for training deep neural networks. Since our coresets approximate every function, they also approximate the gradient of each weight in a neural network, which is a particular function on the input. We then perform empirical evaluations on function approximation and dataset subset selection on popular network architectures and data sets, demonstrating the efficacy and accuracy of our coreset construction.

[ PDF]  [ Poster] 

Author Information

Morad Tukan (DataHeroes)
Samson Zhou (UC Berkeley)
Alaa Maalouf (MIT)
Daniela Rus (MIT CSAIL)
Vladimir Braverman (Johns Hopkins University)
Dan Feldman (The University of Haifa)

More from the Same Authors