Exploring Compositional High Order Pattern Potentials for Structured Output Learning

By Yujia Li, Daniel Tarlow, Richard S. Zemel. 2013

39 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing, Model Architectures, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yoshua Bengio
2. Sergey Levine
3. Quoc V. Le
4. Kyunghyun Cho
5. Mohit Bansal
6. Ruslan Salakhutdinov
7. Trevor Darrell
8. Jianfeng Gao

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Yoshua Bengio
3. Trevor Darrell
4. Dacheng Tao
5. Ming-Hsuan Yang
6. Jitendra Malik
7. Xiaogang Wang
8. Andrew Zisserman

In Natural Language Processing

1. Kyunghyun Cho
2. Yoshua Bengio
3. Ruslan Salakhutdinov
4. Mohit Bansal
5. Jason Aaron Edward Weston
6. Jianfeng Gao
7. Quoc V. Le
8. Mohammad Norouzi

In Model Architectures

1. Yoshua Bengio
2. Ruslan Salakhutdinov
3. Pascal Vincent
4. Mohammad Norouzi
5. Jitendra Malik
6. L. Yao
7. Yizhou Yu
8. Saizheng Zhang

In Signal Processing

1. Yoshua Bengio
2. Pascal Vincent
3. Ruslan Salakhutdinov
4. L. Yao
5. Mohammad Norouzi
6. Saizheng Zhang
7. Jason Yosinski
8. Guillaume Alain

Citations in the corpus

• GSNs : Generative Stochastic Networks (2015, background): “, 2014) or structured output (Li et al., 2013) models? This paper lays theoretical foundations for a move in this direction through the following main contributions:”
• Deep Learning of Representations: Looking Forward (2013, methodology): “, as has been done for structured output predictions when there is complex probabilistic structure between the output variables (Mnih et al., 2011; Li et al., 2013).”
• Iterative Instance Segmentation (2015, background): “Later work considers extending these models by incorporating higher-order potentials of specific forms for which inference is tractable [18, 21].”
• Learning Stochastic Feedforward Neural Networks (2013, methodology): “C-GRBMs are popular models used for human motion modeling [6], structured prediction [10], and as a higher-order potential in image segmentation [11].”
• Combining the Best of Convolutional Layers and Recurrent Layers: A Hybrid Network for Semantic Segmentation (2016, methodology): “MRF, CRF) are then introduced to model the semantic context and obtain a spatially coherent semantic label map [15,16,17,18].”
• Deep Value Networks Learn to Evaluate and Iteratively Refine Structured Outputs (2017, methodology): “Some methods incorporate higher order potentials in CRFs (Kohli et al., 2009) or model global shape priors with Restricted Boltzmann Machines (Li et al., 2013; Kae et al., 2013; Yang et al., 2014; Eslami et al., 2014).”
• Learning Structured Output Representation using Deep Conditional Generative Models (2015, unknown): “In this sense, our work shares a similar motivation to the recent work on image segmentation tasks using hybrid models of CRF and Boltzmann machine [13, 21, 37].”
• Learning High-level Prior with Convolutional Neural Networks for Semantic Segmentation (2015, methodology): “The corresponding qualitative and quantitative assessments are reported and compared to the state-of-art including CRF, CHPOPPs [20], GLOC [14] and MMBM [36].”