- talks - lists.cs.princeton.edu

Nayana Prasad Nagendra will present her Pre FPO "A Frontend For All Workloads" on Monday, November 4, 2019 at 11:30am in CS 302
by Nicki Mahler 28 Oct '19

28 Oct '19

Nayana Prasad Nagendra will present her Pre FPO "A Frontend For All Workloads" on Monday, November 4, 2019 at 11:30am in CS 302. The members of her committee are as follows: David August (adviser), Tipp Moseley (Google), David Wentzlaff (ELE), and Sharad Malik (ELE). Everyone is invited to attend her talk. The abstract follows below. Warehouse-Scale Computing applications experience an order of magnitude more front-end stalls than desktop applications because their code footprints are well beyond the capacity of today’s processor structures, including instruction caches and branch target buffers. Current approaches to reducing front-end stalls rely, at least in part, on some subset of processor structures having sufficient capacity. Since working set sizes for important applications will continue to grow faster than processor structures (27% Year-over-Year growth in web-search instruction working set size for example vs. 7% for L1 instruction cache), these approaches are not viable. Instead, we propose Predicted Path Prefetching with Confidence Hedging, a front-end architecture that does not rely on the capacity to prevent front-end stalls by also prefetching the path not predicted when confidence in the path prediction is low. Initial results show many fewer instruction cache misses per kilo instructions with similar prefetch costs compared to existing techniques.

1 0

Colloquium Speaker Sham Kakade TODAY at 12:30pm
by Emily C. Lawrence 24 Oct '19

24 Oct '19

CS Colloquium Speaker Sham Kakade from University of Washington Thursday, October 24 - 12:30pm Friend Center Convocation Room Host: Sanjeev Arora [ https://www.cs.princeton.edu/events/25891 | https://www.cs.princeton.edu/events/25891 ] Representation, Modeling, and Optimization in Reinforcement Learning Reinforcement learning is now the dominant paradigm for how an agent learns to interact with the world. The approach has led to successes ranging across numerous domains, including game playing and robotics, and it holds much promise in new domains, from self-driving cars to interactive medical applications. Some of the central challenges are: -Representational learning: does having a good representation of the environment permit efficient reinforcement learning? -Modeling: should we explicitly build a model of our environment or, alternatively, should we directly learn how to act? -Optimization: in practice, deployed algorithms often use local search heuristics. Can we provably understand when these approaches are effective and provide faster and more robust alternatives? This talk will survey a number of results on these basic questions. Throughout, we will highlight the interplay of theory, algorithm design, and practice. Bio: Sham Kakade is a Washington Research Foundation Data Science Chair, with a joint appointment in the Department of Computer Science and the Department of Statistics at the University of Washington, and is a co-director for the Algorithmic Foundations of Data Science Institute. He works on the mathematical foundations of machine learning and AI. Sham's thesis helped in laying the foundations of the PAC-MDP framework for reinforcement learning. With his collaborators, his additional contributions include: one of the first provably efficient policy search methods, Conservative Policy Iteration, for reinforcement learning; developing the mathematical foundations for the widely used linear bandit models and the Gaussian process bandit models; the tensor and spectral methodologies for provable estimation of latent variable models (applicable to mixture of Gaussians, HMMs, and LDA); the first sharp analysis of the perturbed gradient descent algorithm, along with the design and analysis of numerous other convex and non-convex algorithms. He is the recipient of the IBM Goldberg best paper award (in 2007) for contributions to fast nearest neighbor search and the best paper, INFORMS Revenue Management and Pricing Section Prize (2014). He has been program chair for COLT 2011. Sham was an undergraduate at Caltech, where he studied physics and worked under the guidance of John Preskill in quantum computing. He then completed his Ph.D. in computational neuroscience at the Gatsby Unit at University College London, under the supervision of Peter Dayan. He was a postdoc at the Dept. of Computer Science, University of Pennsylvania , where he broadened his studies to include computational game theory and economics from the guidance of Michael Kearns. Sham has been a Principal Research Scientist at Microsoft Research, New England, an associate professor at the Department of Statistics, Wharton, UPenn , and an assistant professor at the Toyota Technological Institute at Chicago.

1 0

Reminder: Colloquium Speaker Sham Kakade tomorrow at 12:30pm
by Emily C. Lawrence 23 Oct '19

23 Oct '19

CS Colloquium Speaker Sham Kakade from University of Washington Thursday, October 24 - 12:30pm Friend Center Convocation Room Host: Sanjeev Arora [ https://www.cs.princeton.edu/events/25891 | https://www.cs.princeton.edu/events/25891 ] Representation, Modeling, and Optimization in Reinforcement Learning Reinforcement learning is now the dominant paradigm for how an agent learns to interact with the world. The approach has led to successes ranging across numerous domains, including game playing and robotics, and it holds much promise in new domains, from self-driving cars to interactive medical applications. Some of the central challenges are: -Representational learning: does having a good representation of the environment permit efficient reinforcement learning? -Modeling: should we explicitly build a model of our environment or, alternatively, should we directly learn how to act? -Optimization: in practice, deployed algorithms often use local search heuristics. Can we provably understand when these approaches are effective and provide faster and more robust alternatives? This talk will survey a number of results on these basic questions. Throughout, we will highlight the interplay of theory, algorithm design, and practice. Bio: Sham Kakade is a Washington Research Foundation Data Science Chair, with a joint appointment in the Department of Computer Science and the Department of Statistics at the University of Washington, and is a co-director for the Algorithmic Foundations of Data Science Institute. He works on the mathematical foundations of machine learning and AI. Sham's thesis helped in laying the foundations of the PAC-MDP framework for reinforcement learning. With his collaborators, his additional contributions include: one of the first provably efficient policy search methods, Conservative Policy Iteration, for reinforcement learning; developing the mathematical foundations for the widely used linear bandit models and the Gaussian process bandit models; the tensor and spectral methodologies for provable estimation of latent variable models (applicable to mixture of Gaussians, HMMs, and LDA); the first sharp analysis of the perturbed gradient descent algorithm, along with the design and analysis of numerous other convex and non-convex algorithms. He is the recipient of the IBM Goldberg best paper award (in 2007) for contributions to fast nearest neighbor search and the best paper, INFORMS Revenue Management and Pricing Section Prize (2014). He has been program chair for COLT 2011. Sham was an undergraduate at Caltech, where he studied physics and worked under the guidance of John Preskill in quantum computing. He then completed his Ph.D. in computational neuroscience at the Gatsby Unit at University College London, under the supervision of Peter Dayan. He was a postdoc at the Dept. of Computer Science, University of Pennsylvania , where he broadened his studies to include computational game theory and economics from the guidance of Michael Kearns. Sham has been a Principal Research Scientist at Microsoft Research, New England, an associate professor at the Department of Statistics, Wharton, UPenn , and an assistant professor at the Toyota Technological Institute at Chicago.

1 0

Dingli Yu will present his General Exam on Thursday, October 24, 2019 at 9:30am in CS 401.
by Nicki Mahler 22 Oct '19

22 Oct '19

Dingli Yu will present his General Exam on Thursday, October 24, 2019 at 9:30am in CS 401. The members of his committee are as follows: Sanjeev Arora (adviser), Jason D. Lee (ELE), and Jia Deng. Everyone is invited to attend his talk, and those faculty wishing to remain for the oral exam following are welcome to do so. His abstract and reading list follow below. Title: Harnessing the Power of Infinitely Wide Deep Nets Abstract: Recent research shows that the following two models are equivalent: (a) infinitely wide neural networks (NNs) trained under l2 loss by gradient descent with infinitesimally small learning rate (b) kernel regression with respect to so-called Neural Tangent Kernels (NTKs) (Jacot et al., 2018). An efficient algorithm to compute the NTK, as well as its convolutional counterparts (CNTK), appears in Arora et al. (2019a), yielded the finding that classification accuracy of CNTK on CIFAR-10 is within 6-7% of that of the corresponding CNN architecture (best figure being around 78%). Here, we show that NTK and CNTK or their enhanced version perform strongly in the following tasks. 1. On a standard testbed of classification/regression tasks from the UCI database, NTK SVM beats the previous gold standard, Random Forests (RF), and also the corresponding finite nets. 2. On VOC07 testbed for few-shot image classification tasks on ImageNet with transfer learning (Goyal et al., 2019), replacing the linear SVM currently used with a Convolutional NTK SVM consistently improves performance. 3. On CIFAR-10 with 10 – 640 training samples, CNTK consistently beats ResNet-34 by 1% - 3%. On the whole CIFAR-10 dataset, CNTK achieves 89% accuracy, matching the performance of AlexNet (Krizhevsky et al., 2012) by incorporating horizontal flip data augmentation, a pre-processing technique proposed by Coates et al. (2011) and a new operation called Local Average Pooling (LAP) which preserves efficient computability of the kernel and inherits the spirit of standard data augmentation using pixel shifts. Reading list: 1. Mehryar Mohri, Afshin Rostamizadeh, and Ameet Talwalkar. Foundations of Machine Learning. MIT Press, 2018 2. Peter L Bartlett, and Shahar Mendelson, Rademacher and Gaussian complexities: Risk bounds and structural results. Journal of Machine Learning Research, 3(Nov):463–482, 2002. 3. Sanjeev Arora, Simon S Du, Wei Hu, Zhiyuan Li, Ruslan Salakhutdinov, and Ruosong Wang. On exact computation with an infinitely wide neural net. arXiv preprint arXiv:1904.11955, 2019. 4. Sanjeev Arora, Simon S Du, Wei Hu, Zhiyuan Li, and Ruosong Wang. Fine-grained analysis of optimization and generalization for overparameterized two-layer neural networks. arXiv preprint arXiv:1901.08584, 2019b. 5. Simon S. Du, Xiyu Zhai, Barnabas Poczos, and Aarti Singh. Gradient descent provably optimizes over-parameterized neural networks. In International Conference on Learning Representations, 2019. 6. Simon S Du, Jason D Lee, Haochuan Li, LiweiWang, and Xiyu Zhai. Gradient descent finds global minima of deep neural networks. arXiv preprint arXiv:1811.03804, 2018. 7. Benjamin Recht, Rebecca Roelofs, Ludwig Schmidt, and Vaishaal Shankar. Do imagenet classifiers generalize to imagenet? arXiv preprint arXiv:1902.10811, 2019. 8. Matthew Olson, Abraham Wyner, and Richard Berk. Modern neural networks generalize on small data sets. In Advances in Neural Information Processing Systems, pp. 3619-3628, 2018. 9. Manuel Fernandez-Delgado, Eva Cernadas, Senen Barro, and Dinani Amorim. Do we need hundreds of classifiers to solve real world classification problems? The Journal of Machine Learning Research, 15(1):3133-3181, 2014. 10. Julien Mairal, Piotr Koniusz, Zaid Harchaoui, and Cordelia Schmid. Convolutional kernel networks.In Advances in neural information processing systems, pages 2627–2635, 2014. 11. Adam Coates, Andrew Ng, and Honglak Lee. An analysis of single-layer networks in unsupervised feature learning. In Proceedings of the fourteenth international conference on artificial intelligence and statistics, pp. 215–223, 2011. 12. Shuxiao Chen, Edgar Dobriban, and Jane H Lee. Invariance reduces variance: Understanding data augmentation in deep learning and beyond. arXiv preprint arXiv:1907.10905, 2019.

1 0

Nikunj Saunshi will present his General Exam on Wednesday, October 23, 2019 at 9:15am in CS 402
by Nicki Mahler 21 Oct '19

21 Oct '19

Nikunj Saunshi will present his General Exam on Wednesday, October 23, 2019 at 9:15am in CS 402. The members of his committee are Sanjeev Arora (adviser), Elad Hazan, and Karthik Narasimhan. Everyone is invited to attend his talk, and those faculty wishing to remain for the oral exam following are welcome to do so. His abstract and reading list follow below. Title: A theoretical framework to analyze contrastive learning methods for representation learning Abstract: Contrastive learning, reminiscent of word2vec, is a family of methods that use unlabeled data to learn data representations, — e.g., sentence embeddings — that are useful for downstream classification tasks. Training leverages access to semantically similar pairs of data points and encourages high inner product for representations of similar pairs and low inner product for random pairs of data points. We provide a theoretical framework for understanding such methods by positing that similar pairs come from the same latent class. Using a within-class independent sampling assumption, we show that small unsupervised loss in contrastive learning guarantees good performance on linear classification tasks involving these latent classes. This novel connection between the unlabeled data distribution and downstream tasks also helps us show (labeled) sample complexity benefits of such representation learning methods. We then extend this framework by leveraging a graph-theoretic view of the unlabeled data distribution that motivates a more relaxed and realistic data model for similar pairs: cross subclass independence (CSI). Under this new assumption, we show similar guarantees for downstream classification performance by employing a duality-based analysis. Using the graph formulation, we also highlight the need for incorporating assumptions about inductive biases of the representation function class in order to show guarantees under even weaker assumptions Reading list ========================================= 1. Shalev-Shwartz, Shai, and Shai Ben-David. Understanding machine learning: From theory to algorithms. Cambridge University Press, 2014. 2. Logeswaran, L. and Lee, H. An efficient framework for learning sentence representations. In Proceedings of the International Conference on Learning Representations, 2018 3. Mikolov, T., Sutskever, I., Chen, K., Corrado, G. S., and Dean, J. Distributed representations of words and phrases and their compositionality. In Neural Information Processing Systems, 2013. 4. Arora, S., Li, Y., Liang, Y., Ma, T. and Andrej Risteski. A latent variable model approach to PMI-based word embeddings. Transaction of Association for Computational Linguistics, 2016. 5. Gutmann, M. and Hyvarinen, A. Noise-contrastive estimation: A new estimation principle for unnormalized statistical models. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, pp. 297–304, 2010. 6. Ma, Z. and Collins, M. Noise contrastive estimation and negative sampling for conditional models: Consistency and statistical efficiency. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing 7. Maurer, A., Pontil, M., and Romera-Paredes, B. The benefit of multitask representation learning. J. Mach. Learn. Res., 2016. 8. Syed, U. and Schapire, R. A reduction from apprenticeship learning to classification. In Advances in Neural Information Processing Systems 2010. 9. Ross, S., Gordon, G. and Bagnell, D. A reduction of imitation learning and structured prediction to no-regret online learning. In Proceedings of the fourteenth international conference on artificial intelligence and statistics 2011. 10. Sun, W., Vemula, A., Boots, B. and Bagnell, D.. Provably Efficient Imitation Learning from Observation Alone. Proceedings of the 36th International Conference on Machine Learning, 2019. ==========================================

1 0

Colloquium Speaker Andrés Monroy-Hernández, TODAY at 12:30pm
by Emily C. Lawrence 21 Oct '19

21 Oct '19

CS Colloquium Speaker Andrés Monroy-Hernández from Snap Inc. Monday, October 21 - 12:30pm Computer Science - Room 105 Host: Adam Finkelstein [ https://www.cs.princeton.edu/events/25890 | https://www.cs.princeton.edu/events/25890 ] Designing and deploying social computing systems inside and outside the lab Social computing has permeated most aspects of our lives, from work to play. In recent years, however, social platforms have faced challenges ranging from labor concerns about crowd work to misinformation on social media. These and other challenges, along with the emergence of new technical platforms, create opportunities to re-imagine the future of the field. In this talk, I will discuss my research designing and deploying social computing systems to help people connect and collaborate to learn new skills, crowdsource news reporting, and delegate tasks to hybrid AI systems. First, I describe an online community I created to connect millions of young creators to learn computational thinking skills as they create, share, and remix games and animations on the web. Then I shift to how this work inspired my next set of systems to connect urban residents as they share news about their local communities. Finally, I will discuss a novel workplace productivity tool we created to enable information workers to delegate tasks to hybrid intelligence systems powered by humans and AI. I close by articulating the challenges and opportunities ahead for the field, and the ways my team is beginning to explore new systems that support more authentic and intimate connections, building on new technologies including wearables sensors and AR. Bio: Andrés Monroy-Hernández is a lead research scientist at Snap Inc , where he manages the human-computing interaction research team. He is also an affiliate professor at the University of Washington. His work focuses on the design and study of social computing technologies that help people connect and collaborate. Previously, he was at Microsoft Research for six years, where he led the research and development of Calendar.help , Microsoft’s first hybrid AI product. His research has received best paper awards at CHI, CSCW, HCOMP, and ICWSM, and has been featured in The New York Times, CNN, Wired, BBC, and The Economist. Andrés was named one of the 35 Innovators under 35 by the MIT Technology Review magazine in Latin America, and one the most influential Latinos in Tech by CNET. Andrés holds master’s and Ph.D. from MIT, where he led the creation of the Scratch online community. He has a BS from Tec de Monterrey in México. More info at [ http://andresmh.com/ | http://andresmh.com ]

1 0

Reminder: Colloquium Speaker Andrés Monroy-Hernández, tomorrow at 12:30pm
by Emily C. Lawrence 20 Oct '19

20 Oct '19

CS Colloquium Speaker Andrés Monroy-Hernández from Snap Inc. Monday, October 21 - 12:30pm Computer Science - Room 105 Host: Adam Finkelstein [ https://www.cs.princeton.edu/events/25890 | https://www.cs.princeton.edu/events/25890 ] Designing and deploying social computing systems inside and outside the lab Social computing has permeated most aspects of our lives, from work to play. In recent years, however, social platforms have faced challenges ranging from labor concerns about crowd work to misinformation on social media. These and other challenges, along with the emergence of new technical platforms, create opportunities to re-imagine the future of the field. In this talk, I will discuss my research designing and deploying social computing systems to help people connect and collaborate to learn new skills, crowdsource news reporting, and delegate tasks to hybrid AI systems. First, I describe an online community I created to connect millions of young creators to learn computational thinking skills as they create, share, and remix games and animations on the web. Then I shift to how this work inspired my next set of systems to connect urban residents as they share news about their local communities. Finally, I will discuss a novel workplace productivity tool we created to enable information workers to delegate tasks to hybrid intelligence systems powered by humans and AI. I close by articulating the challenges and opportunities ahead for the field, and the ways my team is beginning to explore new systems that support more authentic and intimate connections, building on new technologies including wearables sensors and AR. Bio: Andrés Monroy-Hernández is a lead research scientist at Snap Inc , where he manages the human-computing interaction research team. He is also an affiliate professor at the University of Washington. His work focuses on the design and study of social computing technologies that help people connect and collaborate. Previously, he was at Microsoft Research for six years, where he led the research and development of Calendar.help , Microsoft’s first hybrid AI product. His research has received best paper awards at CHI, CSCW, HCOMP, and ICWSM, and has been featured in The New York Times, CNN, Wired, BBC, and The Economist. Andrés was named one of the 35 Innovators under 35 by the MIT Technology Review magazine in Latin America, and one the most influential Latinos in Tech by CNET. Andrés holds master’s and Ph.D. from MIT, where he led the creation of the Scratch online community. He has a BS from Tec de Monterrey in México. More info at [ http://andresmh.com/ | http://andresmh.com ]

1 0

Colloquium Speaker Sham Kakade Thursday, October 24 at 12:30pm
by Emily C. Lawrence 18 Oct '19

18 Oct '19

CS Colloquium Speaker Sham Kakade from University of Washington Thursday, October 24 - 12:30pm Friend Center Convocation Room Host: Sanjeev Arora [ https://www.cs.princeton.edu/events/25891 | https://www.cs.princeton.edu/events/25891 ] Representation, Modeling, and Optimization in Reinforcement Learning Reinforcement learning is now the dominant paradigm for how an agent learns to interact with the world. The approach has led to successes ranging across numerous domains, including game playing and robotics, and it holds much promise in new domains, from self-driving cars to interactive medical applications. Some of the central challenges are: -Representational learning: does having a good representation of the environment permit efficient reinforcement learning? -Modeling: should we explicitly build a model of our environment or, alternatively, should we directly learn how to act? -Optimization: in practice, deployed algorithms often use local search heuristics. Can we provably understand when these approaches are effective and provide faster and more robust alternatives? This talk will survey a number of results on these basic questions. Throughout, we will highlight the interplay of theory, algorithm design, and practice. Bio: Sham Kakade is a Washington Research Foundation Data Science Chair, with a joint appointment in the Department of Computer Science and the Department of Statistics at the University of Washington, and is a co-director for the Algorithmic Foundations of Data Science Institute. He works on the mathematical foundations of machine learning and AI. Sham's thesis helped in laying the foundations of the PAC-MDP framework for reinforcement learning. With his collaborators, his additional contributions include: one of the first provably efficient policy search methods, Conservative Policy Iteration, for reinforcement learning; developing the mathematical foundations for the widely used linear bandit models and the Gaussian process bandit models; the tensor and spectral methodologies for provable estimation of latent variable models (applicable to mixture of Gaussians, HMMs, and LDA); the first sharp analysis of the perturbed gradient descent algorithm, along with the design and analysis of numerous other convex and non-convex algorithms. He is the recipient of the IBM Goldberg best paper award (in 2007) for contributions to fast nearest neighbor search and the best paper, INFORMS Revenue Management and Pricing Section Prize (2014). He has been program chair for COLT 2011. Sham was an undergraduate at Caltech, where he studied physics and worked under the guidance of John Preskill in quantum computing. He then completed his Ph.D. in computational neuroscience at the Gatsby Unit at University College London, under the supervision of Peter Dayan. He was a postdoc at the Dept. of Computer Science, University of Pennsylvania , where he broadened his studies to include computational game theory and economics from the guidance of Michael Kearns. Sham has been a Principal Research Scientist at Microsoft Research, New England, an associate professor at the Department of Statistics, Wharton, UPenn , and an assistant professor at the Toyota Technological Institute at Chicago.

1 0

Colloquium Speaker Ramón Cáceres TODAY at 1:30pm
by Emily C. Lawrence 16 Oct '19

16 Oct '19

Distinguished CS Colloquium Speaker Ramón Cáceres from Google Wednesday, October 16 - 1:30pm Computer Science - Room 105 Host: Margaret Martonosi [ https://www.cs.princeton.edu/events/25878 | https://www.cs.princeton.edu/events/25878 ] Zanzibar: Google’s Consistent, Global Authorization System Determining whether online users are authorized to access digital objects is central to preserving privacy. This talk presents the design, implementation, and deployment of Zanzibar, a global system for storing and evaluating access control lists. Zanzibar provides a uniform data model and configuration language for expressing a wide range of access control policies from hundreds of client services at Google, including Calendar, Cloud, Drive, Maps, Photos, and YouTube. Its authorization decisions respect causal ordering of user actions and thus provide external consistency amid changes to access control lists and object contents. Zanzibar scales to trillions of access control lists and millions of authorization requests per second to support services used by billions of people. It has maintained 95th-percentile latency of less than 10 milliseconds and availability of greater than 99.999% over 3 years of production use. Bio: Ramón Cáceres is a software engineer at Google. He was previously a researcher at AT&T Labs and IBM Research. His areas of focus include computer systems and networks, mobile computing services and applications, and location data analysis and modeling. He holds a PhD in Computer Science from the University of California at Berkeley. He is an IEEE Fellow and an ACM Distinguished Scientist.

1 0

Colloquium Speaker Dougal Maclaurin TODAY at 12:30
by Emily C. Lawrence 15 Oct '19

15 Oct '19

CS Colloquium Speaker Dougal Maclaurin from Google Brain Tuesday, October 15 - 12:30pm Computer Science - Room 105 Host: Ryan Adams [ https://www.cs.princeton.edu/events/25877 | https://www.cs.princeton.edu/events/25877 ] JAX: Accelerated machine learning research via composable function transformations in Python JAX is a system for high-performance machine learning research. It offers the familiarity of Python+NumPy and the speed of hardware accelerators, and it enables the definition and the composition of function transformations useful for machine learning programs. In particular, these transformations include automatic differentiation, automatic batching, end-to-end-compilation (via XLA), and parallelizing over multiple accelerators. They are the key to JAX's power and to its relative simplicity. JAX had its initial open-source release in December 2018 (https://github.com/google/jax). It is currently being used by several groups of researchers for a wide range of advanced applications, from studying spectra of neural networks, to probabilistic programming and Monte Carlo methods, and scientific applications in physics and biology. Users appreciate JAX most of all for its ease of use and flexibility. Bio: Dougal Maclaurin is a research scientist at Google. He works on programming languages and systems for machine learning, particularly the Python library JAX. He started Autograd , a system for automatic differentiation in Python, which has inspired the design of several systems, including PyTorch , MinPy , Torch Autograd and Julia Autograd . He is a co-founder of Day Zero Diagnostics, a startup developing a sequencing-based diagnostic for drug-resistant infections. He received his Ph.D. from Harvard in 2016, working with Ryan Adams on the development of methods for machine learning. His work on scalable MCMC, "Firefly Monte Carlo", was recognized with the Best Paper award at UAI 2014.

1 0