End-to-end learning potentials for structured attribute prediction

نویسندگان

  • Kota Yamaguchi
  • Takayuki Okatani
  • Takayuki Umeda
  • Kazuhiko Murasaki
  • Kyoko Sudo
چکیده

We present a structured inference approach in deep neural networks for multiple attribute prediction. In attribute prediction, a common approach is to learn independent classifiers on top of a good feature representation. However, such classifiers assume conditional independence on features and do not explicitly consider the dependency between attributes in the inference process. We propose to formulate attribute prediction in terms of marginal inference in the conditional random field. We model potential functions by deep neural networks and apply the sum-product algorithm to solve for the approximate marginal distribution in feed-forward networks. Our message passing layer implements sparse pairwise potentials by a softplus-linear function that is equivalent to a higher-order classifier, and learns all the model parameters by end-to-end back propagation. The experimental results using SUN attributes and CelebA datasets suggest that the structured inference improves the attribute prediction performance, and possibly uncovers the hidden relationship between attributes.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Learning Arbitrary Potentials in CRFs with Gradient Descent

Are we using the right potential functions in the Condi-tional Random Field models that are popular in the Visioncommunity? Semantic segmentation and other pixel-levellabelling tasks have made significant progress recently dueto the deep learning paradigm. However, most state-of-the-art structured prediction methods also include a randomfield model with a hand-crafted Ga...

متن کامل

End-to-End Learning for Structured Prediction Energy Networks

Structured Prediction Energy Networks (Belanger & McCallum, 2016) (SPENs) are a simple, yet expressive family of structured prediction models. An energy function over candidate structured outputs is given by a deep network, and predictions are formed by gradient-based optimization. Unfortunately, we have struggled to apply the structured SVM (SSVM) learning method of Belanger & McCallum (2016) ...

متن کامل

Learning to Search: Structured Prediction Techniques for Imitation Learning

Modern robots successfully manipulate objects, navigate rugged terrain, drive in urban settings, and play world-class chess. Unfortunately, programming these robots is challenging, timeconsuming and expensive; the parameters governing their behavior are often unintuitive, even when the desired behavior is clear and easily demonstrated. Inspired by successful end-to-end learning systems such as ...

متن کامل

Adversarial Deep Structured Nets for Mass Segmentation from Mammograms

Mass segmentation provides effective morphological features which are important for mass diagnosis. In this work, we propose a novel end-to-end network for mammographic mass segmentation which employs a fully convolutional network (FCN) to model a potential function, followed by a CRF to perform structured learning. Because the mass distribution varies greatly with pixel position, the FCN is co...

متن کامل

Fast, Exact and Multi-scale Inference for Semantic Image Segmentation with Deep Gaussian CRFs

In this work we propose a combination of the Gaussian Conditional Random Field (G-CRF) with Deep Learning for the task of structured prediction. Our method inherits several virtues of G-CRF and Deep Learning: (a) the structured prediction task has a unique global optimum that is obtained exactly from the solution of a linear system (b) structured prediction can be jointly trained in an end-to-e...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • CoRR

دوره abs/1708.01892  شماره 

صفحات  -

تاریخ انتشار 2017