
Efficient and Modular Implicit Differentiation
Automatic differentiation (autodiff) has revolutionized machine learning...
read it

Decomposing reversemode automatic differentiation
We decompose reversemode automatic differentiation into (forwardmode) ...
read it

The advantages of multiple classes for reducing overfitting from test set reuse
Excessive reuse of holdout data can lead to overfitting. However, there ...
read it

Measuring the Effects of Data Parallelism on Neural Network Training
Recent hardware developments have made unprecedented amounts of data par...
read it

Estimation from Indirect Supervision with Linear Moments
In structured prediction problems where we have indirect supervision of ...
read it

Principal Component Projection Without Principal Component Analysis
We show how to efficiently project a vector onto the top principal compo...
read it

Toward Deeper Understanding of Neural Networks: The Power of Initialization and a Dual View on Expressivity
We develop a general duality between neural networks and compositional k...
read it

Unregularizing: approximate proximal point and faster stochastic algorithms for empirical risk minimization
We develop a family of accelerated stochastic algorithms that minimize s...
read it

Competing with the Empirical Risk Minimizer in a Single Pass
In many estimation problems, e.g. linear and logistic regression, we wis...
read it

Relaxations for inference in restricted Boltzmann machines
We propose a relaxationbased approximate inference algorithm that sampl...
read it
Roy Frostig
is this you? claim profile