Daphne Koller

Support vector machine active learning with applications to text classification

Support vector machines have met with significant success in numerous real-world learning tasks. However, like most machine learning algorithms, they are generally applied using a randomly selected training set classified in advance. In many settings, we also have the option of using pool-based active learning. Instead of using a randomly selected training set, the learner has access to a pool of unlabeled instances and can request the labels for some number of them. We introduce a new algorithm for performing active learning with support vector machines, i.e., an algorithm for choosing which instances to request next. We provide a theoretical motivation for the algorithm using the notion of a version space. We present experimental results showing that employing our active learning method can significantly reduce the need for labeled training instances in both the standard inductive and transductive settings.

Population genomics of human gene expression

Genetic variation influences gene expression, and this variation in gene expression can be efficiently mapped to specific genomic regions and variants. Here we have used gene expression profiling of Epstein-Barr virus–transformed lymphoblastoid cell lines of all 270 individuals genotyped in the HapMap Consortium to elucidate the detailed features of genetic variation underlying gene expression variation. We find that gene expression is heritable and that differentiation between populations is in agreement with earlier small-scale studies. A detailed association analysis of over 2.2 million common SNPs per population (5% frequency in HapMap) with gene expression identified at least 1,348 genes with association signals in cis and at least 180 in trans. Replication in at least one independent population was achieved for 37% of cis signals and 15% of trans signals, respectively. Our results strongly support an abundance of cis-regulatory variation in the human genome. Detection of trans effects is limited but suggests that regulatory variation may be the key primary effect contributing to phenotypic variation in humans. We also explore several methodologies that improve the current state of analysis of gene expression variation.

SCAPE: shape completion and animation of people

We introduce the SCAPE method (Shape Completion and Animation for PEople)---a data-driven method for building a human shape model that spans variation in both subject shape and pose. The method is based on a representation that incorporates both articulated and non-rigid deformations. We learn a pose deformation model that derives the non-rigid surface deformation as a function of the pose of the articulated skeleton. We also learn a separate model of variation based on body shape. Our two models can be combined to produce 3D surface models with realistic muscle deformation for different people in different poses, when neither appear in the training set. We show how the model can be used for shape completion --- generating a complete surface mesh given a limited set of markers specifying the target shape. We present applications of shape completion to partial view completion and motion capture animation. In particular, our method is capable of constructing a high-quality animated surface model of a moving person, with realistic muscle deformation, using just a single static scan and a marker motion capture sequence of the person.

The Chemical Genomic Portrait of Yeast: Uncovering a Phenotype for All Genes

Genetics aims to understand the relation between genotype and phenotype. However, because complete deletion of most yeast genes (∼80%) has no obvious phenotypic consequence in rich medium, it is difficult to study their functions. To uncover phenotypes for this nonessential fraction of the genome, we performed 1144 chemical genomic assays on the yeast whole-genome heterozygous and homozygous deletion collections and quantified the growth fitness of each deletion strain in the presence of chemical or environmental stress conditions. We found that 97% of gene deletions exhibited a measurable growth phenotype, suggesting that nearly all genes are essential for optimal growth in at least one condition.

A module map showing conditional activity of expression modules in cancer

DNA microarrays are widely used to study changes in gene expression in tumors, but such studies are typically system-specific and do not address the commonalities and variations between different types of tumor. Here we present an integrated analysis of 1,975 published microarrays spanning 22 tumor types. We describe expression profiles in different tumors in terms of the behavior of modules, sets of genes that act in concert to carry out a specific function. Using a simple unified analysis, we extract modules and characterize gene-expression profiles in tumors as a combination of activated and deactivated modules. Activation of some modules is specific to particular types of tumor; for example, a growth-inhibitory module is specifically repressed in acute lymphoblastic leukemias and may underlie the deregulated proliferation in these cancers. Other modules are shared across a diverse set of clinical conditions, suggestive of common tumor progression mechanisms. For example, the bone osteoblastic module spans a variety of tumor types and includes both secreted growth factors and their receptors. Our findings suggest that there is a single mechanism for both primary tumor proliferation and metastasis to bone. Our analysis presents multiple research directions for diagnostic, prognostic and therapeutic studies.

Being Bayesian About Network Structure. A Bayesian Approach to Structure Discovery in Bayesian Networks

In many multivariate domains, we are interested in analyzing the dependency structure of the underlying distribution, e.g., whether two variables are in direct interaction. We can represent dependency structures using Bayesian network models. To analyze a given data set, Bayesian model selection attempts to find the most likely (MAP) model, and uses its structure to answer these questions. However, when the amount of available data is modest, there might be many models that have non-negligible posterior. Thus, we want compute the Bayesian posterior of a feature, i.e., the total posterior probability of all models that contain it. In this paper, we propose a new approach for this task. We first show how to efficiently compute a sum over the exponential number of networks that are consistent with a fixed order over network variables. This allows us to compute, for a given order, both the marginal probability of the data and the posterior of a feature. We then use this result as the basis for an algorithm that approximates the Bayesian posterior of a feature. Our approach uses a Markov Chain Monte Carlo (MCMC) method, but over orders rather than over network structures. The space of orders is smaller and more regular than the space of structures, and has much a smoother posterior “landscape”. We present empirical results on synthetic and real-life datasets that compare our approach to full model averaging (when possible), to MCMC over network structures, and to a non-Bayesian bootstrap approach.

Simultaneous Localization and Mapping with Sparse Extended Information Filters

In this paper we describe a scalable algorithm for the simultaneous mapping and localization (SLAM) problem. SLAM is the problem of acquiring a map of a static environment with a mobile robot. The vast majority of SLAM algorithms are based on the extended Kalman filter (EKF). In this paper we advocate an algorithm that relies on the dual of the EKF, the extended information filter (EIF). We show that when represented in the information form, map posteriors are dominated by a small number of links that tie together nearby features in the map. This insight is developed into a sparse variant of the EIF, called the sparse extended information filter (SEIF). SEIFs represent maps by graphical networks of features that are locally interconnected, where links represent relative information between pairs of nearby features, as well as information about the robot’s pose relative to the map. We show that all essential update equations in SEIFs can be executed in constant time, irrespective of the size of the map. We also provide empirical results obtained for a benchmark data set collected in an outdoor environment, and using a multi-robot mapping simulation.

Decomposing a scene into geometric and semantically consistent regions

High-level, or holistic, scene understanding involves reasoning about objects, regions, and the 3D relationships between them. This requires a representation above the level of pixels that can be endowed with high-level attributes such as class of object/region, its orientation, and (rough 3D) location within the scene. Towards this goal, we propose a region-based model which combines appearance and scene geometry to automatically decompose a scene into semantically meaningful regions. Our model is defined in terms of a unified energy function over scene appearance and structure. We show how this energy function can be learned from data and present an efficient inference technique that makes use of multiple over-segmentations of the image to propose moves in the energy-space. We show, experimentally, that our method achieves state-of-the-art performance on the tasks of both multi-class image segmentation and geometric reasoning. Finally, by understanding region classes and geometry, we show how our model can be used as the basis for 3D reconstruction of the scene.

Towards robust automatic traffic scene analysis in real-time

Automatic symbolic traffic scene analysis is essential to many areas of IVHS (Intelligent Vehicle Highway Systems). Traffic scene information can be used to optimize traffic flow during busy periods, identify stalled vehicles and accidents, and aid the decision-making of an autonomous vehicle controller. Improvements in technologies for machine vision-based surveillance and high-level symbolic reasoning have enabled the authors to develop a system for detailed, reliable traffic scene analysis. The machine vision component of the system employs a contour tracker and an affine motion model based on Kalman filters to extract vehicle trajectories over a sequence of traffic scene images. The symbolic reasoning component uses a dynamic belief network to make inferences about traffic events such as vehicle lane changes and stalls. In this paper, the authors discuss the key tasks of the vision and reasoning components as well as their integration into a working prototype. Preliminary results of an implementation on special purpose hardware using C-40 Digital Signal Processors show that near real-time performance can be achieved without further improvements.

Learning structured prediction models: a large margin approach

We consider large margin estimation in a broad range of prediction models where inference involves solving combinatorial optimization problems, for example, weighted graph-cuts or matchings. Our goal is to learn parameters such that inference using the model reproduces correct answers on the training data. Our method relies on the expressive power of convex optimization problems to compactly capture inference or solution optimality in structured prediction models. Directly embedding this structure within the learning formulation produces concise convex problems for efficient estimation of very complex and diverse models. We describe experimental results on a matching task, disulfide connectivity prediction, showing significant improvements over state-of-the-art methods.