DeepAI AI Chat
Log In Sign Up

Notes on Latent Structure Models and SPIGOT

by   André F. T. Martins, et al.
Unbabel Inc.

These notes aim to shed light on the recently proposed structured projected intermediate gradient optimization technique (SPIGOT, Peng et al., 2018). SPIGOT is a variant of the straight-through estimator (Bengio et al., 2013) which bypasses gradients of the argmax function by back-propagating a surrogate "gradient." We provide a new interpretation to the proposed gradient and put this technique into perspective, linking it to other methods for training neural networks with discrete latent variables. As a by-product, we suggest alternate variants of SPIGOT which will be further explored in future work.


page 1

page 2

page 3

page 4


Coupled Gradient Estimators for Discrete Latent Variables

Training models with discrete latent variables is challenging due to the...

Understanding the Mechanics of SPIGOT: Surrogate Gradients for Latent Structure Learning

Latent structure models are a powerful tool for modeling language data: ...

Notes on asymptotics of sample eigenstructure for spiked covariance models with non-Gaussian data

These expository notes serve as a reference for an accompanying post Mor...

Backpropagating through Structured Argmax using a SPIGOT

We introduce the structured projection of intermediate gradients optimiz...

Straight-Through Estimator as Projected Wasserstein Gradient Flow

The Straight-Through (ST) estimator is a widely used technique for back-...

Likelihood-free inference with an improved cross-entropy estimator

We extend recent work (Brehmer, et. al., 2018) that use neural networks ...

Eigenvalue-corrected Natural Gradient Based on a New Approximation

Using second-order optimization methods for training deep neural network...