WorldCat Identities

Koller, Daphne

Works: 86 works in 123 publications in 2 languages and 920 library holdings
Genres: Educational films  Internet videos  Conference papers and proceedings 
Roles: Author, Thesis advisor
Publication Timeline
Most widely held works about Daphne Koller
Most widely held works by Daphne Koller
Probabilistic graphical models : principles and techniques by Daphne Koller( Book )

15 editions published between 2009 and 2012 in English and held by 514 WorldCat member libraries worldwide

Proceedings of the annual Conference on Uncertainty in Artificial Intelligence, available for 1991-present. Since 1985, the Conference on Uncertainty in Artificial Intelligence (UAI) has been the primary international forum for exchanging results on the use of principled uncertain-reasoning methods in intelligent systems. The UAI Proceedings have become a basic reference for researches and practitioners who want to know about both theoretical advances and the latest applied developments in the field
TEDTalks : Daphne Koller - What We're Learning from Online Education( Visual )

1 edition published in 2012 in English and held by 242 WorldCat member libraries worldwide

Educator Daphne Koller is enticing top universities to put their most intriguing courses online for free - not just as a service, but as a way to research how people learn. In this TEDTalk, Koller explains how Coursera, a social entrepreneurship company cofounded with Andrew Ng, tracks each keystroke, quiz, peer-to-peer discussion, and self-graded assignment to build an unprecedented pool of data on how knowledge is processed
Uncertainty in artificial intelligence : proceedings of the seventeenth conference (2001), August 2-5, 2001, University of Washington, Seattle, Washington by Jack Breese( Book )

6 editions published in 2001 in English and held by 39 WorldCat member libraries worldwide

From knowledge to belief by Daphne Koller( Book )

7 editions published between 1993 and 1994 in English and held by 11 WorldCat member libraries worldwide

We use techniques from finite model theory to analyze the computational aspects of random worlds. The problem of computing degrees of belief is undecidable in general. However, for unary knowledge bases, a tight connection to the principle of maximum entropy often allows us to compute degrees of belief
Adaptive probabilistic networks by Stuart J Russell( Book )

2 editions published in 1994 in English and held by 7 WorldCat member libraries worldwide

Asymptotic conditional probabilities : the unary case by A. J Grove( Book )

4 editions published in 1993 in English and held by 6 WorldCat member libraries worldwide

Abstract: "Motivated by problems that arise in computing degrees of belief, we consider the problem of computing asymptotic conditional probabilities for first-order sentences. Given first-order sentences [symbol] and [theta], we consider the structures with domain [1 ..., N] that satisfy [theta], and compute the fraction of them in which [symbol] is true. We then consider what happens to this fraction as N gets large. This extends the work on 0-1 laws that considers the limiting probability of first-order sentences, by considering asymptotic conditional probabilities. As shown in [Lio69, GHK93], in the general case, asymptotic conditional probabilities do not always exist, and most questions relating to this issue are highly undecidable. These results, however, all depend on the assumption that [theta] can use a nonunary predicate symbol. Liogon'kiĭ [Lio69] shows that if we condition on formulas [theta] involving unary predicate symbols only (but no equality or constant symbols), then the asymptotic conditional probability does exist and can be effectively computed. This is the case even if we place no corresponding restrictions on [symbol]. We extend this result here to the case where [theta] involves equality and constants. We show that the complexity of computing the limit depends on various factors, such as the depth of quantifier nesting, or whether the vocabulary is finite or infinite. We completely characterize the complexity of the problem in the different cases, and show related results for the associated approximation problem."
Representation dependence in probabilistic inference by Joseph Y Halpern( Book )

2 editions published in 1995 in English and held by 6 WorldCat member libraries worldwide

Abstract: "Non-deductive reasoning systems are often representation dependent: representing the same situation in two different ways may cause such a system to return two different answers. This is generally viewed as a significant problem. For example, the principle of maximum entropy has been subjected to much criticism due to its representation dependence. There has, however, been almost no work investigating representation dependence. In this paper, we formalize this notion and show that it is not a problem specific to maximum entropy. In fact, we show that any probabilistic inference system that sanctions certain important patterns of reasoning, such as minimal default assumption of independence, must suffer from representation dependence. We then show that invariance under a restricted class of representation changes can form a reasonable compromise between representation dependence and other desiderata."
On the complexity of two-person zero-sum games in extensive form by Daphne Koller( Book )

2 editions published in 1990 in English and held by 3 WorldCat member libraries worldwide

Random worlds and maximum entropy by A. J Grove( Book )

2 editions published in 1994 in English and held by 3 WorldCat member libraries worldwide

Abstract: "Given a knowledge base KB containing first-order and statistical facts, we consider a principled method, called the random- worlds method, for computing a degree of belief that some formula [symbol] holds given KB. If the domain has size N, then we can consider all possible worlds, or first-order models, with domain [1 ..., N] that satisfy KB, and compute the fraction of them in which [symbol] is true. We define the degree of belief to be the asymptotic value of this fraction as N grows large. We show that when the vocabulary underlying [symbol] and KB uses constants and unary predicates only, we can naturally associate an entropy with each world. As N grows larger, there are many more worlds with higher entropy. Therefore, we can use a maximum entropy computation to compute the degree of belief. This result is in a similar spirit to previous work in physics (e.g., [Jay78]) and artificial intelligence (e.g., [PV89, Sha89]), but is far more general. Of equal interest to the actual results themselves are the numerous subtle issues we must address when formulating it. For languages with binary predicate symbols, the random-worlds method continues to make sense, but there no longer seems to be any useful connection to maximum entropy. It is difficult to see how maximum entropy can be applied at all. In fact, results from [GHK93a] show that even generalizations of maximum entropy are unlikely to be useful. These observations suggest unexpected limitations to the applicability of maximum entropy methods."
Unraveling the genetics of disease using structured probabilistic models by Alexis Jane Battle( )

1 edition published in 2013 in English and held by 2 WorldCat member libraries worldwide

Recent technological advances have allowed us to collect genomic data on an unprecedented scale, with the promise of revealing genetic variants, genes, and pathways disrupted in clinically relevant human traits. However, identifying functional variants and ultimately unraveling the genetics of complex disease from such data have presented significant challenges. With millions of genetic factors to consider, spurious associations and lack of statistical power are major hurdles. Further, we cannot easily assess functional roles even for known trait-associated variants, particularly for those that lie outside of protein-coding regions of the genome. To address these challenges in identifying the genetic factors underlying complex traits, we have developed probabilistic machine learning methods that leverage biological structure and prior knowledge. In this thesis, we describe four applications of such models. First, we present a method for reconstructing causal gene networks from interventional genetic interaction data in model organisms. Here, we are able to identify intricate functional dependencies among hundreds of genes affecting a complex trait. We have applied this method to understanding the genetics of protein folding in yeast, where we demonstrate ability to recapitulate the details, including ordering, of known pathways, and make novel functional predictions. Second, we present PriorNet, a method for incorporating gene network and path- way information into the analysis of population-level studies of genetic variation in human disease. PriorNet utilizes a flexible, Markov Random Field prior to propagate information between functionally related genes and related diseases, in order to improve statistical power in large-scale disease studies. We demonstrate a significant improvement in the discovery of disease-relevant genes in studies of three autoimmune diseases. Next, we extend the intuitions of PriorNet in a method for identifying interactions between genetic variants in human disease, to begin to understand how genes work together in complex disease processes. Our method, GAIT, leverages gene networks, network structure, and other patterns to adaptively prioritize candidate in- teractions for testing, and dramatically reduce the burden of multiple hypothesis correction to identify a large number of interactions in diverse human disease studies. Finally, we discuss the identification of functional variants on a large scale through the use of gene expression as a high-resolution cellular phenotype. We have sequenced RNA from 922 genotyped individuals to provide a direct window into the distribution, properties, and consequences of thousands of regulatory variants affecting diverse gene expression traits including splicing and allelic expression. From the identified variants, we also train a model, LRVM, for predicting regulatory consequences based on location and genomic properties of each variant
The digital patient : machine learning techniques for analyzing electronic health record data by Suchi Saria( )

1 edition published in 2011 in English and held by 2 WorldCat member libraries worldwide

The current unprecedented rate of digitization of longitudinal health data -- continuous device monitoring data, laboratory measurements, medication orders, treatment reports, reports of physician assessments -- allows visibility into patient health at increasing levels of detail. A clearer lens into this data could help improve decision making both for individual physicians on the front lines of care, and for policy makers setting national direction. However, this type of data is high-dimensional (an infant with no prior clinical history can have more than 1000 different measurements in the ICU), highly unstructured (the measurements occur irregularly, and different numbers and types of measurements are taken for different patients) and heterogeneous (from ultrasound assessments to lab tests to continuous monitor data). Furthermore, the data is often sparse, systematically not present, and the underlying system is non-stationary. Extracting the full value of the existing data requires novel approaches. In this thesis, we develop novel methods to show how longitudinal health data contained in Electronic Health Records (EHRs) can be harnessed for making novel clinical discoveries. For this, one requires access to patient outcome data -- which patient has which complications. We present a method for automated extraction of patient outcomes from EHR data; our method shows how natural languages cues from the physicians notes can be combined with clinical events that occur during a patient's length of stay in the hospital to extract significantly higher quality annotations than previous state-of-the-art systems. We develop novel methods for exploratory analysis and structure discovery in bedside monitor data. This data forms the bulk of the data collected on any patient yet, it is not utilized in any substantive way post collection. We present methods to discover recurring shape and dynamic signatures in this data. While we primarily focus on clinical time series, our methods also generalize to other continuous-valued time series data. Our analysis of the bedside monitor data led us to a novel use of this data for risk prediction in infants. Using features automatically extracted from physiologic signals collected in the first 3 hours of life, we develop Physiscore, a tool that predicts infants at risk for major complications downstream. Physiscore is both fully automated and significantly more accurate than the current standard of care. It can be used for resource optimization within a NICU, managing infant transport to a higher level of care and parental counseling. Overall, this thesis illustrates how the use of machine learning for analyzing these large scale digital patient data repositories can yield new clinical discoveries and potentially useful tools for improving patient care
Alignment of cryo-electron tomography images using Markov Random Fields by Fernando Amat Gil( )

1 edition published in 2010 in English and held by 2 WorldCat member libraries worldwide

Cryo-Electron tomography (CET) is the only imaging technology capable of visualizing the 3D organization of intact bacterial whole cells at nanometer resolution in situ. However, quantitative image analysis of CET datasets is extremely challenging due to very low signal to noise ratio (well below 0dB), missing data and heterogeneity of biological structures. In this thesis, we present a probabilistic framework to align CET images in order to improve resolution and create structural models of different biological structures. The alignment problem of 2D and 3D CET images is cast as a Markov Random Field (MRF), where each node in the graph represents a landmark in the image. We connect pairs of nodes based on local spatial correlations and we find the "best'' correspondence between the two graphs. In this correspondence problem, the "best'' solution maximizes the probability score in the MRF. This probability is the product of singleton potentials that measure image similarity between nodes and the pairwise potentials that measure deformations between edges. Well-known approximate inference algorithms such as Loopy Belief Propagation (LBP) are used to obtain the "best'' solution. We present results in two specific applications: automatic alignment of tilt series using fiducial markers and subtomogram alignment. In the first case we present RAPTOR, which is being used in several labs to enable real high-throughput tomography. In the second case our approach is able to reach the contrast transfer function limit in low SNR samples from whole cells as well as revealing atomic resolution details invisible to the naked eye through nanogold labeling
Efficient computation of equilibria for extensive two-person games by Daphne Koller( Book )

1 edition published in 1994 in English and held by 2 WorldCat member libraries worldwide

Abstract: "The Nash equilibria of a two-person, non-zero-sum game are the solutions of a certain linear complementarity problem (LCP). In order to use this for solving a game in extensive form, it is first necessary to convert the game to a strategic description such as the normal form. The classical normal form, however, is often exponentially large in the size of the game tree. Hence, finding equilibria of extensive games typically implies exponential blowup in terms of both time and space. In this paper we suggest an alternative approach, based on the sequence form of the game. For a game with perfect recall, the sequence form is a linear sized strategic description, which results in an LCP of linear size. For this LCP, we show that an equilibrium can be found efficiently by Lemke's algorithm, a generalization of the Lemke-Howson method."
Computational pathology for genomic medicine by Andrew Hanno Beck( )

1 edition published in 2013 in English and held by 2 WorldCat member libraries worldwide

The medical specialty of pathology is focused on the transformation of information extracted from patient tissue samples into biologically informative and clinically useful diagnoses to guide research and clinical care. Since the mid-19th century, the primary data type used by surgical pathologists has been microscopic images of hematoxylin and eosin stained tissue sections. Over the past several decades, molecular data have been increasingly incorporated into pathological diagnoses. There is now a need for the development of new computational methods to systematically model and integrate these complex data to support the development of data-driven diagnostics for pathology. The overall goal of this dissertation is to develop and apply methods in this new field of Computational Pathology, which is aimed at: 1) The extraction of comprehensive integrated sets of data characterizing disease from a patient's tissue sample; and 2) The application of machine learning-based methods to inform the interpretation of a patient's disease state. The dissertation is centered on three projects, aimed at the development and application of methods in Computational Pathology for the analysis of three primary data types used in cancer diagnostics: 1) morphology; 2) biomarker expression; and 3) genomic signatures. First, we developed the Computational Pathologist (C-Path) system for the quantitative analysis of cancer morphology from microscopic images. We used the system to build a microscopic image-based prognostic model in breast cancer. The C-Path prognostic model outperformed competing approaches and uncovered the prognostic significance of several novel characteristics of breast cancer morphology. Second, to systematically evaluate the biological informativeness and clinical utility of the two most commonly used protein biomarkers (estrogen receptor (ER) and progesterone receptor (PR)) in breast cancer diagnostics, we performed an integrative analysis over publically available expression profiling data, clinical data, and immunohistochemistry data collected from over 4,000 breast cancer patients, extracted from 20 published studies. We validated our findings on an independent integrated breast cancer dataset from over 2,000 breast cancer patients in the Nurses' Health Study. Our analyses demonstrated that the ER-/PR+ disease subtype is rare and non-reproducible. Further, in our genomewide study we identified hundreds of biomarkers more informative than PR for the stratification of both ER+ and ER- disease. Third, we developed a new computational method, Significance Analysis of Prognostic Signatures (SAPS), for the identification of robust prognostic signatures from clinically annotated Omics data. We applied SAPS to publically available clinically annotated gene expression data obtained from over 3,800 breast cancer patients from 19 published studies and over 1,700 ovarian cancer patients from 11 published studies. Using these two large meta-datasets, we applied SAPS and performed the largest analysis of subtype-specific prognostic pathways ever performed in breast or ovarian cancer. Our analyses led to the identification of a core set of prognostic biological signatures in breast and ovarian cancer and their molecular subtypes. Further, the SAPS method should be generally useful for future studies aimed at the identification of biologically informative and clinically useful signatures from clinically annotated Omics data. Taken together, these studies provide new insights into the biological factors driving cancer progression, and our methods and models will support the continuing development of the field of Computational Pathology
Gai shuai tu mo xing : Yuan li yu ji shu by Daphne Koller( Book )

2 editions published in 2015 in Chinese and held by 2 WorldCat member libraries worldwide

Ben shu xiang xi lun shu le you xiang tu mo xing he wu xiang tu mo xing de biao shi,Tui li he xue xi wen ti,Quan er zong jie le ren gong zhi neng zhei yi qian yan yan jiu ling yu de zui xin jin zhan.Wei le bian yu du zhe li jie,Shu zhong bao han le da liang de ding yi,Ding li,Zheng ming,Suan fa ji qi wei dai ma,Chuan tong le da liang de fu zhu cai liao,Ru shi li(examples),Ji qiao zhuan lan(skill boxes),Shi li zhuan lan(case studyboxes),Gai nian zhuan lan(concept boxes)Deng.Ling wai,Zai di 2 zhang jie shao le gai lü lun he tu lun de he xin zhi shi,Zai fu lu zhong jie shao le xin xi lun,Suan fa fu za xing,Zu he you hua deng bu chong cai liao,Wei xue xi he yun yong gai lü tu mo xing ti gong le wan bei de ji chu
Sensors & Symbols: An Integrated Framework( Book )

2 editions published in 1999 in English and held by 1 WorldCat member library worldwide

The goal of this effort was to provide a unified probabilistic framework that integrates symbolic and sensory reasoning. Such a framework would allow sensor data to be analyzed in terms of high-level symbolic models. It will also allow the results of high-level analysis to guide the low-level sensor interpretation task and to help in resolving ambiguities in the sensor data. Our approach was based on the framework of probabilistic graphical models, which allows us to build systems that learn and reason with complex models, encompassing both low-level continuous sensor data and high-level symbolic concepts. Over the five years of the project, we explored two main thrusts: Inference and learning in hybrid and temporal Bayesian networks Mapping and modeling of 3D physical environments. Our progress on each of these two directions is detailed in the attached report
Visual learning with weakly labeled video by Kevin Tang( )

1 edition published in 2015 in English and held by 1 WorldCat member library worldwide

With the rising popularity of Internet photo and video sharing sites like Flickr, Instagram, and YouTube, there is a large amount of visual data uploaded to the Internet on a daily basis. In addition to pixels, these images and videos are often tagged with the visual concepts and activities they contain, leading to a natural source of weakly labeled visual data, in which we aren't told where within the images and videos these concepts or activities occur. By developing methods that can effectively utilize weakly labeled visual data for tasks that have traditionally required clean data with laborious annotations, we can take advantage of the abundance and diversity of visual data on the Internet. In the first part of this thesis, we consider the problem of complex event recognition in weakly labeled video. In weakly labeled videos, it is often the case that the complex events we are interested in are not temporally localized, and the videos contain varying amounts of contextual or unrelated segments. In addition, the complex events themselves often vary significantly in the actions they consist of, as well as the sequences in which they occur. To address this, we formulate a flexible, discriminative model that is able to learn the latent temporal structure of complex events from weakly labeled videos, resulting in a better understanding of the complex events and improved recognition performance. The second part of this thesis tackles the problem of object localization in weakly labeled video. Towards this end, we focus on several aspects of the object localization problem. First, using object detectors trained from images, we formulate a method for adapting these detectors to work well in video data by discovering and adapting them to examples automatically extracted from weakly labeled videos. Then, we explore separately the use of large amounts of negative and positive weakly labeled visual data for object localization. With only negative weakly labeled videos that do not contain a particular visual concept, we show how a very simple metric allows us to perform distributed object segmentation in potentially noisy, weakly labeled videos. With only positive weakly labeled images and videos that share a common visual concept, we show how we can leverage correspondence information between images and videos to identify and detect the common object. Lastly, we consider the problem of learning temporal embeddings from weakly labeled video. Using the implicit weak label that videos are sequences of temporally and semantically coherent images, we learn temporal embeddings for frames of video by associating frames with the temporal context that they appear in. These embeddings are able to capture semantic context, which results in better performance for a wide variety of standard tasks in video
Knowledge Representation for an Uncertain World( Book )

2 editions published in 1997 in English and held by 1 WorldCat member library worldwide

Any application where an intelligent agent interacts with the real world must deal with the problem of uncertainty. Bayesian belief networks have become dominant in addressing this issue. This is a framework based on principled probabilistic semantics, which achieves effective knowledge representation and inference capabilities by utilizing the locality structure in the domain: typically, only very few aspects of the situation directly affect each other. Despite their success, belief networks are inadequate as a knowledge representation language for large, complex domains: Their attribute-based nature does not allow us to express general rules that hold in many different circumstances. This prevents knowledge from being shared among applications; the initial knowledge acquisition cost has to be paid for each new domain. It also inhibits the construction of large complex networks. We deal with this issue by presenting a rich knowledge-representation language from which belief networks can be constructed to suit specific circumstances, algorithms for learning the network parameters from data, fast approximate inference algorithms designed to deal with the large networks that result. We show how these techniques can be applied in domains involving continuous variables, in situations where the world changes over time, and in the context of planing under uncertainty
Probabilistic models for region-based scene understanding by Stephen Gould( )

1 edition published in 2010 in English and held by 1 WorldCat member library worldwide

One of the long-term goals of computer vision is to be able to understand the world through visual images. This daunting task involves reasoning simultaneously about objects, regions and 3D geometry. Traditionally, computer vision research has tackled these tasks is isolation: independent detectors for finding objects, image segmentation algorithms for defining regions, and specialized monocular depth perception methods for reconstructing geometry. Unfortunately, this isolated reasoning can lead to inconsistent interpretations of the scene. In this thesis we develop a unified probabilistic model that avoids these inconsistencies. We introduce a region-based representation of the scene in which pixels are grouped together to form consistent regions. Each region is then annotated with a semantic and geometric class label. Next, we extend our representation to include the concept of objects, which can be comprised of multiple regions. Finally, we show how our region-based representation can be used to interpret the 3D structure of the scene. Importantly, we model the scene using a coherent probabilistic model over random variables defined by our region-based representation. This enforces consistency between tasks and allows contextual dependencies to be modeled across tasks, e.g., that sky should be above the horizon, and ground below it. Finally, we present an efficient algorithm for performing inference in our model, and demonstrate state-of-the-art results on a number of standard tasks
Making Rational Decisions Using Adaptive Utility Elicitation( )

1 edition published in 2000 in Undetermined and held by 1 WorldCat member library worldwide

moreShow More Titles
fewerShow Fewer Titles
Audience Level
Audience Level
  Kids General Special  
Audience level: 0.55 (from 0.19 for TEDTalks : ... to 0.95 for Sensors & ...)

Probabilistic graphical models : principles and techniques
Alternative Names
Dafne Kollere

Daphne Koller amerykańska informatyczka

Daphne Koller informatica statunitense

Daphne Koller informatica uit Israël

Daphne Koller informaticienne américaine

Daphne Koller profesora estadounidense de informática

Daphne Koller US-amerikanische Informatikerin

Koller, Daphne

Дафна Коллер

Дафни Колер

Դաֆնա Կոլլեր

דפנה קולר

דפנה קולר מהנדסת ישראלית

دافنی کالر

دافنی کالەر



Uncertainty in artificial intelligence : proceedings of the seventeenth conference (2001), August 2-5, 2001, University of Washington, Seattle, Washington