Corpus ID: 1791473

OptNet: Differentiable Optimization as a Layer in Neural Networks

@inproceedings{Amos2017OptNetDO,
  title={OptNet: Differentiable Optimization as a Layer in Neural Networks},
  author={Brandon Amos and J. Z. Kolter},
  booktitle={ICML},
  year={2017}
}
  • Brandon Amos, J. Z. Kolter
  • Published in ICML 2017
  • Computer Science, Mathematics
  • This paper presents OptNet, a network architecture that integrates optimization problems (here, specifically in the form of quadratic programs) as individual layers in larger end-to-end trainable deep networks. These layers encode constraints and complex dependencies between the hidden states that traditional convolutional and fully-connected layers often cannot capture. In this paper, we explore the foundations for such an architecture: we show how techniques from sensitivity analysis, bilevel… CONTINUE READING
    Differentiable Convex Optimization Layers
    45
    Differentiable Fixed-Point Iteration Layer
    1
    Physarum Powered Differentiable Linear Programming Layers and Applications
    CNNS THROUGH DIFFERENTIABLE PDE LAYER
    Differentiable Forward and Backward Fixed-Point Iteration Layers.
    1
    Differentiable Learning of Submodular Models
    25
    Homogeneous Linear Inequality Constraints for Neural Network Activations
    1
    End-to-End Learnable Geometric Vision by Backpropagating PnP Optimization
    1
    MIPaaL: Mixed Integer Program as a Layer
    12

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 42 REFERENCES
    Input Convex Neural Networks
    73
    Adam: A Method for Stochastic Optimization
    46811
    Conditional Random Fields as Recurrent Neural Networks
    1803
    Learning Deep Structured Models
    189
    On Differentiating Parameterized Argmin and Argmax Problems with Application to Bi-level Optimization
    64
    Generative Adversarial Nets
    16734
    Generic Methods for Optimization-Based Modeling
    155
    End-to-End Learning for Structured Prediction Energy Networks
    79