Results 1 -
8 of
8
Structured Output Learning with High Order Loss Functions
"... Often when modeling structured domains, it is desirable to leverage information that is not naturally expressed as simply a label. Examples include knowledge about the evaluation measure that will be used at test time, and partial (weak) label information. When the additional information has structu ..."
Abstract
-
Cited by 16 (4 self)
- Add to MetaCart
(Show Context)
Often when modeling structured domains, it is desirable to leverage information that is not naturally expressed as simply a label. Examples include knowledge about the evaluation measure that will be used at test time, and partial (weak) label information. When the additional information has structure that factorizes according to small subsets of variables (i.e., is low order, or decomposable), several approaches can be used to incorporate it into a learning procedure. Our focus in this work is the more challenging case, where the additional information does not factorize according to low order graphical model structure; we call this the high order case. We propose to formalize various forms of this additional information as high order loss functions, which may have complex interactions over large subsets of variables. We then address the computational challenges inherent in learning according to such loss functions, particularly focusing on the loss-augmented inference problem that arises in large margin learning; we show that learning with high order loss functions is often practical, giving strong empirical results, with one popular and several novel high-order loss functions, in several settings. 1
Fast exact inference for recursive cardinality models
- In Uncertainty in Artificial Intelligence
, 2012
"... Cardinality potentials are a generally useful class of high order potential that affect probabilities based on how many of D binary variables are active. Maximum a posteriori (MAP) inference for cardinality potential models is well-understood, with efficient computations taking O(D log D) time. Yet ..."
Abstract
-
Cited by 8 (4 self)
- Add to MetaCart
Cardinality potentials are a generally useful class of high order potential that affect probabilities based on how many of D binary variables are active. Maximum a posteriori (MAP) inference for cardinality potential models is well-understood, with efficient computations taking O(D log D) time. Yet efficient marginalization and sampling have not been addressed as thoroughly in the machine learning community. We show that there exists a simple algorithm for computing marginal probabilities and drawing exact joint samples that runs in O(D log 2 D) time, and we show how to frame the algorithm as efficient belief propagation in a low order tree-structured model that includes additional auxiliary variables. We then develop a new, more general class of models, termed Recursive Cardinality models, which take advantage of this efficiency. Finally, we show how to do efficient exact inference in models composed of a tree structure and a cardinality potential. We explore the expressive power of Recursive Cardinality models and empirically demonstrate their utility. 1
unknown title
"... Visual scene segmentation is the partitioning and parsing of a visual scene into distinct coherent parts, regions or surfaces, separated by boundaries or surface discontinuities. This is a fundamental process in visual perceptual organization that is important for downstream inference of object surf ..."
Abstract
- Add to MetaCart
(Show Context)
Visual scene segmentation is the partitioning and parsing of a visual scene into distinct coherent parts, regions or surfaces, separated by boundaries or surface discontinuities. This is a fundamental process in visual perceptual organization that is important for downstream inference of object surface geometries, shapes, and identities. Computationally, this problem is very challenging and has engaged computer vision
Neural encoding of scene statistics for surface and object inference
"... Features associated with an object or its surfaces in natural scenes tend to vary coherently in space and time. In psychological literature, these coherent covariations have been considered to be important for neural systems to acquire models of objects and object categories. From a statistical infe ..."
Abstract
- Add to MetaCart
(Show Context)
Features associated with an object or its surfaces in natural scenes tend to vary coherently in space and time. In psychological literature, these coherent covariations have been considered to be important for neural systems to acquire models of objects and object categories. From a statistical inference perspective, such coherent covariation can provide a mechanism to learn the statistical priors in natural scenes that are useful for probabilistic inference. In this article, we present some neurophysiological experimental observations in the early visual cortex that provide insights on how correlation structures in visual scenes are being encoded by neuronal tuning and connections among neurons. The key insight is that correlated structures in visual scenes result in correlated neuronal activities, which shapes the tuning properties of individual neurons and the connections between them, embedding Gestalt-related computational constraints or priors for surface inference. Extending these concepts to the inferotemporal cortex suggests a representational framework that is distinct from traditional feed-forward hierarchy of invariant object representation and recognition. In this framework, lateral connections among view-based neurons, learned from the temporal association of the object views observed over time, can form a linked graph structure with local dependency, akin to a dense aspect graph in computer vision. This web-like graph allows view-invariant object representation to be created using sparse feed-forward connections, while maintaining the explicit representation of the different views. Thus, it can serve as an effective prior model for generating predictions of future incoming views to facilitate object inference.
Multiscale Fields of Patterns
"... We describe a general framework for defining high-order image models that can be used in a variety of applications. The approach involves modeling local patterns in a multiscale representation of an image. Local properties of a coarsened image re-flect non-local properties of the original image. In ..."
Abstract
- Add to MetaCart
(Show Context)
We describe a general framework for defining high-order image models that can be used in a variety of applications. The approach involves modeling local patterns in a multiscale representation of an image. Local properties of a coarsened image re-flect non-local properties of the original image. In the case of binary images local properties are defined by the binary patterns observed over small neighborhoods around each pixel. With the multiscale representation we capture the frequency of patterns observed at different scales of resolution. Our framework leads to expres-sive priors that depend on a relatively small number of parameters. For inference and learning we use an MCMC method for block sampling with very large blocks. We evaluate the approach with two example applications. One involves contour detection. The other involves binary segmentation. 1
Published In Neural encoding of scene statistics for surface and object inference
"... Neural encoding of scene statistics for surface and object inference ..."
(Show Context)
A Class of Compression Systems with Model-Free Encoding
"... Abstract — Practical compression systems are constrained by their bit-stream standards, which define the source model to-gether with the coding method used. We introduce a model-free coding architecture that separates the two aspects of compression and allows the design of potentially more powerful ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract — Practical compression systems are constrained by their bit-stream standards, which define the source model to-gether with the coding method used. We introduce a model-free coding architecture that separates the two aspects of compression and allows the design of potentially more powerful source models, as well as more flexible use of the compressed information stream. We show that this architecture is capable of producing competitive performance while supporting new use cases. I.
Multiscale Belief Propagation on Concrete CT Image Fast Segmentation
"... An image segmentation fast method based on multi-scale belief propagation is proposed to solve the concrete CT image segmentations problem. Firstly, according to the feature of belief propagation algorithm, a self-characteristic multiscale belief propagation(MBP) is proposed; Then, according to comp ..."
Abstract
- Add to MetaCart
An image segmentation fast method based on multi-scale belief propagation is proposed to solve the concrete CT image segmentations problem. Firstly, according to the feature of belief propagation algorithm, a self-characteristic multiscale belief propagation(MBP) is proposed; Then, according to compute complexity problem in process of belief messages propagation, a method to reduce quantity of algorithm compute is proposed; Finally, using standard images to validate nicety and speediness on our method,and applying on concrete CT image segmentation. The experiment results show that the proposed method can improve efficiency and precision of the image segmentation, and afford an important assisting method on concrete meso-structure CT image study of architecture projection. The method has important projection applying meaning.
