- talks - lists.cs.princeton.edu

Mihir Kulkarni will present his general exam "Training models for NLP tasks on novel domains using transfer learning" on Wednesday, August 7, 2019 at 3pm in Friend 006
by Nicki Mahler 01 Aug '19

01 Aug '19

Mihir Kulkarni will present his general exam "Training models for NLP tasks on novel domains using transfer learning" on Wednesday, August 7, 2019 at 3pm in Friend 006. The members of his committee are as follows: Karthik Narasimhan (advisor), Sanjeev Arora, and Olga Russakovsky. 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. Abstract: Recent advances in machine learning for NLP have shown that there are great benefits to unsupervised pre-training followed by finetuning to specific tasks (BERT, GPT): in particular, BERT was able to obtain state of the art results on 11 NLP tasks using this method. However, having these models perform well on domains that are different from the one where they were trained is still a challenge. In our work, we explore methods to transfer representations learnt by BERT to different models for improved performance on NLP tasks in a different domain-- medical papers. We generate a dataset from medical reviews obtainable on the Cochrane Database of Systematic Reviews, to train models to learn representations on technical documents that match representations learnt by BERT on plain text summaries of those documents. We experiment with different methods to perform section-wise matching between documents to create our parallel corpus, both for performing the matching and transferring representations. Performance on the domain transfer task is reported on both intrinsic and extrinsic evaluation metrics including named entity recognition on the EBM-NLP and sentence similarity on the MED-STS datasets. Reading list: 1. Ian Goodfellow, Yoshua Bengio, Aaron Courville, Deep Learning, MIT Press, 2016. 2. Devlin, J., Chang, M.W., Lee, K. and Toutanova, K., 2018. BERT: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. 3. Peters, Matthew E. and Neumann, Mark and Iyyer, Mohit and Gardner, Matt and Clark, Christopher and Lee, Kenton and Zettlemoyer, Luke: Deep contextualized word representations, Proc. of NAACL, 2018. 4. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł. and Polosukhin, I: Attention is all you need. In Advances in neural information processing systems, 2017. 5. Jiaqi Mu, Suma Bhat, Pramod Viswanath: All-but-the-Top: Simple and effective Postprocessing for Word Representations, International Conference on Learning Representations, 2018. 6. Tan, Chuanqi & Sun, Fuchun & Kong, Tao & Zhang, Wenchang & Yang, Chao & Liu, Chunfang: A Survey on Deep Transfer Learning: 27th International Conference on Artificial Neural Networks, Rhodes, Greece, October 4–7, 2018, Proceedings, Part III. 10.1007/978-3-030-01424-7_27. 7. Sinno Jialin Pan and Qiang Yang and Wei Fan and Sinno Jialin Pan: A survey on transfer learning, IEEE Transactions on Knowledge and Data Engineering, 2010. 8. Yifan Peng and Shankai Yan and Zhiyong Lu, Transfer Learning in Biomedical Natural Language Processing: An Evaluation of BERT and ELMo on Ten Benchmarking Datasets, Proceedings of the 2019 Workshop on Biomedical Natural Language Processing (BioNLP 2019), 2019. 9.Benjamin Nye, Junyi Jessy Li, Roma Patel, Yinfei Yang, Iain J. Marshall, Ani Nenkova, Byron C. Wallace: A Corpus with Multi-Level Annotations of Patients, Interventions and Outcomes to Support Language Processing for Medical Literature, Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 2018. 10. Yanshan Wang, Naveed Afzal, Sunyang Fu, Liwei Wang, Feichen Shen, Majid Rastegar-Mojarad, Hongfang Liu: MedSTS: a resource for clinical semantic textual similarity, Language Resources and Evaluation, 2018.

1 0

Brian Bullins will present his FPO, "Efficient Higher-Order Optimization for Machine Learning" on Tuesday, 7/23/2019 at 2pm in CS 402
by Nicki Mahler 22 Jul '19

22 Jul '19

Brian Bullins will present his FPO, "Efficient Higher-Order Optimization for Machine Learning" on Tuesday, 7/23/2019 at 2pm in CS 402. The members of his committee are: Elad Hazan (Adviser), Examiners: Mark Braverman and Sanjeev Arora; Readers: Yoram Singer and Yuxin Chen (EE). A copy of his thesis, is available upon request. Everyone is invited to attend his talk. The talk abstract follows below: In recent years, stochastic gradient descent (SGD) has taken center stage for training large-scale models in machine learning. Although some higher-order methods have improved iteration complexity in theory, the per-iteration costs render them unusable when faced with millions of parameters and training examples. In this thesis, I will present several works which enable higher-order optimization to be as scalable as first-order methods. The first method is a stochastic second-order algorithm for convex optimization, called LiSSA, which uses Hessian information to construct an unbiased Newton step in time linear in the problem dimension. To bypass the typical eciency barriers for second-order methods, we harness the ERM structure in standard machine learning tasks. While convex problems allow for global convergence, recent state-of-the-art models, such as deep neural networks, highlight the importance of developing a better understanding of non-convex guarantees. In order to handle this challenging setting, I will present FastCubic, a Hessian-based method which provably converges to first-order critical points faster than gradient descent, while additionally converging to second-order critical points. Finally, I will establish how to leverage even higher-order derivative information by means of the FastQuartic algorithm, which achieves faster convergence for both smooth and non-smooth convex optimization problems by combining ecient tensor methods with near-optimal higher-order acceleration.

1 1

Ruth Dannenfelser will present her Pre FPO "From gene expression data to biomedical abstracts: data-driven approaches and systems to interrogate complex disease" on Tuesday, July 16, 2019 at 9am in CS 402.
by Nicki Mahler 11 Jul '19

11 Jul '19

Ruth Dannenfelser will present her Pre FPO "From gene expression data to biomedical abstracts: data-driven approaches and systems to interrogate complex disease" on Tuesday, July 16, 2019 at 9am in CS 402. The members of her committee are: Olga Troyanskaya (adviser), Mona Singh, Kai Li, Yibin Kang (Mol Bio), Wendell Lim (UCSF). Everyone is invited to attend her talk. The talk title and abstract follow below: Title: From gene expression data to biomedical abstracts: data-driven approaches and systems to interrogate complex disease Large-scale genomic studies now give more predictive power than ever, allowing us to profile the composition of tissues, study cellular functions, and understand organismal traits at an unprecedented level of detail. This is particularly important for studying heterogeneous diseases, such as cancer, where small patient-specific differences play critical roles in disease development and progression. Here we build off the wealth of publicly available data to examine the interplay between cancer and the immune system and then develop two query-based visualization systems that enable interactive data exploration for the biomedical community at large. The first part of my talk will present two perspectives on cancer and the immune system, starting with a semi-supervised approach for immune cell type quantification. Using derived immune markers we examine lymphocyte infiltration in breast cancer and find that estrogen receptor activity and genomic complexity are the key factors driving variation in lymphocytic infiltrate across individuals. In a broader scope, we leverage public gene expression data to further the development of targeted immunotherapeutics for solid tumors. Engineered T cell therapies have shown great promise for hematological cancers but have only found limited success in targeting solid tumors due to off target effects. Working closely with experimental collaborators we are developing a method to prioritize safer pairs of antigen targets that will help engineered T cells hone in on specific tumor targets minimizing damage to normal tissues. The second half will cover how we can extract unbiased signals from large collections of biomedical data in the form of scientific abstracts and large repositories of transcriptomics data. First, we show how we can obtain informative tissue-disease-gene relationships from abstracts and integrate them into a web system that presents different snapshots of curated interactions and adds tissue and disease annotations to gene lists from experimental assays (e.g. GWAS, differentially expressed genes, drug screens, etc). Secondly, we extend SEEK, a gene expression search engine that simultaneously returns coexpressed genes and relevant datasets where the genes are likely coregulated. Our extension enables significant performance improvements, expands the search space across the major model organisms, and provides a new cross-organism exploration interface to help facilitate translational research.

1 0

Maciej Halber will present his FPO, "RGBD pipeline for Indoor Scene Reconstruction and Understanding" on Monday, July 15th, 2019 at 1pm in CS 402.
by Nicki Mahler 08 Jul '19

08 Jul '19

Maciej Halber will present his FPO,"RGBD pipeline for Indoor Scene Reconstruction and Understanding" on Monday, July 15th, 2019 at 1pm in CS 402. The members of his committee are Thomas Funkhouser (adviser); Readers: Szymon Rusinkiewicz and Matthias Niessner (The Technical University of Munich); Examiners: Thomas Funkhouser, Olga Russakovsky, and Adam Finkelstein. A copy of his thesis is available upon request. Everyone is invited to attend his talk. The talk abstract follows below: In this work, we consider the problem of reconstructing a 3D model from a sequence of color and depth frames. Generating such a model has many important applications, ranging from the entertainment industry to real estate. However, transforming the RGBD frames into high-quality 3D models is a challenging problem, especially if additional semantic information is required. In this document, we introduce three projects, which implement various stages of a robust RGBD processing pipeline. First, we consider the challenges arising during the RGBD data capture process. While the depth cameras are providing dense, per-pixel depth measurements, there is a non-trivial error associated with the resulting data. We discuss the depth generation problem and propose an error reduction technique based on estimating an image-space undistortion field. We describe the capture process of the data required for the generation of such an undistortion field. We showcase how correcting the depth measurements improves the reconstruction quality. Second, we address the problem of registering RGBD frames over a long video sequence into a globally consistent 3D model. We propose a “fine-to-coarse” global registration algorithm that leverages robust registrations at finer scales to seed detection and enforcement of geometrical constraints, modeled as planar structures, at coarser scales. To test global registration algorithms, we provide a benchmark with 10,401 manually-clicked point correspondences in 25 scenes from the SUN3D dataset. We find that our fine-to-coarse algorithm registers long RGBD sequences better than previous methods. Last, we show how repeated scans of the same space can be used to establish associations between the different observations. Specifically, we consider a situation where 3D scans are acquired repeatedly at sparse time intervals. We develop an algorithm that analyzes these “rescans” and builds a temporal model of a scene with semantic instance information. The proposed algorithm operates inductively by using a temporal model resulting from past observations to infer instance segmentation of a new scan. The temporal model is continuously updated to reflect the changes that occur in the scene over time, providing object associations across time. The algorithm outperforms alternate approaches based on state-of-the-art networks for semantic instance segmentation.

1 0

Elena Balashova will present her FPO, "Shape Synthesis Using Structure-Aware Reasoning" on Monday, July 8, 2019 at 1pm in CS 105
by Nicki Mahler 08 Jul '19

08 Jul '19

Elena Balashova will present her FPO, "Shape Synthesis Using Structure-Aware Reasoning" on Monday, July 8, 2019 at 1pm in CS 105. The members of her committee are as follows: Thomas Funkhouser (adviser), Readers: Szymon Rusinkiewicz and Vivek Singh (Siemens Healthcare); Examiners: Olga Russakovsky, Adam Finkelstein, and Thomas Funkhouser (adviser). A copy of her thesis is available upon request. Everyone is invited to attend her talk. The talk abstract follows below. Shape synthesis is an important area of computer vision and graphics that concerns creation of new shapes and reconstruction from partial data. Its goal is to learn a model that can generate shapes within an object category suitable for novel shape creation, interpolation, completion, editing, and other geometric modeling applications. Existing tools learn shape properties from large collections of shapes. Although these methods have been very successful at learning how to synthesize the coarse shapes of objects in categories with highly diverse shapes, they have not always produced examples that reconstruct important structural elements of a shape. In this thesis, I describe how structure can be incorporated into the synthesis process, and how it can be used to improve generative models. First, I introduce a template-defined skeleton structure for learning a part-aware generative model in typography, where the shapes have a known structure and can be explained by a small number of templates. Next, I present a scenario of noisy archaeological wall painting (fresco) reconstruction from eroded fragments, where there is no well-defined structure and exponentially many arrangement possibilities in this case, I present a cluster evaluation function that guides the assembly process and encourages selection of good clusters. Finally, I describe a semantic landmark-based structure and how it can be used to improve a generative model of examples with extremely varied topology by means of a geometric shape-structure consistency loss. Through exploration of each type of structure, I show how reasoning with proposed structures helps synthesize more accurate and realistic shapes. I also propose a fully automatic framework for font completion. Finally, I design a genetic algorithm for wall painting reconstruction and propose an iterative outlier detection technique based on the eigenvector method.

1 1

EE Seminar - June 13, 2019, 11:00 am, E-Quad B205 - Prof. Zhenkai Liang
by Emily Lawrence 10 Jun '19

10 Jun '19

Department of Electrical Engineering Seminar Speaker: Prof. Zhenkai Liang, National University of Singapore Title: Scaling up Binary Analysis via Knowledge-oriented Techniques Date: Thursday, June 13, 2019 Time: 11:00 am Room: E-Quad, Room B205 Host: Prof. Ruby Lee Abstract: Binary analysis is a fundamental technique in software and system security. It has a wide range of applications, such as vulnerability discovery, attack response, malware analysis, and software testing and debugging. Due to the lack of high-level semantics and complex program behaviors, it is challenging for binary analysis solutions to scale up to large real-world binaries in practice. Existing solutions are often task-driven and bounded by a practical time limit, which hinders comprehensively understanding of binaries. Furthermore, it is difficult to integrate the knowledge generated from different solutions. In this talk, we discuss our research in scaling up binary analysis in a knowledge-oriented manner. We believe knowledge abstraction is the key to scale up binary analysis, where binary analysis solutions generate understandings that can be shared and reused in other solution. Our investigation includes techniques for knowledge extraction, tools for knowledge integration, and platforms for knowledge accumulations and sharing. The accumulated knowledge not only allows broader and deeper analysis into binaries. It also enables emerging data-driven and learning techniques to be effectively adopted in binary analysis solutions. In this talk, I will also share our experience and reflection in system security education. Bio: Zhenkai Liang is an Associate Professor of the School of Computing, National University of Singapore. His main research interests are in system and software security, web security, mobile security, and program analysis. He is also the Co-Lead PI of National Cybersecurity R&D Lab in Singapore. He has served as the technical program committee members of many system security conferences, including the ACM Conference on Computer and Communications Security (CCS), USENIX Security Symposium and the Network and Distributed System Security Symposium (NDSS), as well as a member of NDSS Steering Group. As a co-author, he received the Best Paper Award in ICECCS 2014, the Best Paper Award in W2SP 2014, the ACM SIGSOFT Distinguished Paper Award at ESEC/FSE 2009, the Best Paper Award at USENIX Security Symposium 2007, and the Outstanding Paper Award at ACSAC 2003. He also won the Annual Teaching Excellence Award of National University of Singapore in 2014 and 2015. He received his Ph.D. degree in Computer Science from Stony Brook University in 2006, and B.S. degrees in Computer Science and Economics from Peking University in 1999.

1 0

Yiying Zhang TODAY at 1:30pm
by Emily C. Lawrence 21 May '19

21 May '19

Yiying Zhang from Purdue University Tuesday, May 21, 2019 Computer Science - Room 105 at 1:30pm Host: Amit Levy [ https://www.cs.princeton.edu/events/25862 | https://www.cs.princeton.edu/events/25862 ] Farewell to Servers: Software and Hardware Approaches towards Datacenter Resource Disaggregation Datacenters have been using the "monolithic" server model for decades, where each server hosts a set of hardware devices like CPU and DRAM on a motherboard and runs an OS on top to manage the hardware resources. This monolithic server model fundamentally restricts datacenters from achieving efficient resource packing, hardware rightsizing, and great heterogeneity. Recent hardware and application trends such as serverless computing further call for a rethinking of the long-standing server-centric model. My answer is to "disaggregate" monolithic servers into network-attached hardware components that host different hardware resources and offer different functionalities (e.g., a processor component for computation, a memory component for fast data accesses). I believe that after evolving from physical (DC-1.0) to virtual (DC-2.0), datacenters should evolve further into a disaggregated one (DC-3.0), where hardware resources can be allocated and scaled to the exact amount that applications use and can be individually managed and customized for different application needs. By not having servers, DC-3.0 disrupts designs and technologies in almost every layer in today's datacenters, from hardware and networking to OS and applications. My lab undertook pioneering efforts in building an end-to-end solution for DC-3.0 with a new OS, a new hardware platform, and a new network system. This talk will focus on two systems that are central to the design of DC-3.0: 1) LegoOS , a new distributed operating system designed for managing disaggregated resources. LegoOS splits OS functionalities into different units, each running at a hardware component and managing the component's hardware resources. LegoOS enables the disaggregation and customization of OS functionalities, a significant step towards building DC-3.0's software infrastructure. 2) LegoFPGA , a new approach of using FPGA to efficiently manage and virtualize hardware resources. LegoFPGA offers a solution to co-design application, OS, and hardware functionalities and customize them for different hardware resources and application domains, an important step towards building DC-3.0's hardware infrastructure. With LegoOS and LegoFPGA , we demonstrate that separating core OS and hardware functionalities is not only feasible but can largely improve performance per dollar over the current datacenter monolithic server model. Bio: Yiying Zhang is an assistant professor in the School of Electrical and Computer Engineering at Purdue University. Her research interests span operating systems, distributed systems, computer architecture, and datacenter networking. She also works on the intersection of systems and programming language, security, and AI/ML. She won an OSDI best paper award in 2018 and an NSF CAREER award in 2019. Yiying’s lab is among the few groups in the world now that build new OSes and full-stack, cross-layer systems. Yiying received her Ph.D. from the Department of Computer Sciences at the University of Wisconsin-Madison under the supervision of Andrea and Remzi Arpaci-Dusseau and worked as a postdoctoral scholar at the University of California, San Diego before joining Purdue.

1 0

Reminder: Yiying Zhang tomorrow at 1:30pm
by Emily C. Lawrence 20 May '19

20 May '19

Yiying Zhang from Purdue University Tuesday, May 21, 2019 Computer Science - Room 105 at 1:30pm Host: Amit Levy [ https://www.cs.princeton.edu/events/25862 | https://www.cs.princeton.edu/events/25862 ] Farewell to Servers: Software and Hardware Approaches towards Datacenter Resource Disaggregation Datacenters have been using the "monolithic" server model for decades, where each server hosts a set of hardware devices like CPU and DRAM on a motherboard and runs an OS on top to manage the hardware resources. This monolithic server model fundamentally restricts datacenters from achieving efficient resource packing, hardware rightsizing, and great heterogeneity. Recent hardware and application trends such as serverless computing further call for a rethinking of the long-standing server-centric model. My answer is to "disaggregate" monolithic servers into network-attached hardware components that host different hardware resources and offer different functionalities (e.g., a processor component for computation, a memory component for fast data accesses). I believe that after evolving from physical (DC-1.0) to virtual (DC-2.0), datacenters should evolve further into a disaggregated one (DC-3.0), where hardware resources can be allocated and scaled to the exact amount that applications use and can be individually managed and customized for different application needs. By not having servers, DC-3.0 disrupts designs and technologies in almost every layer in today's datacenters, from hardware and networking to OS and applications. My lab undertook pioneering efforts in building an end-to-end solution for DC-3.0 with a new OS, a new hardware platform, and a new network system. This talk will focus on two systems that are central to the design of DC-3.0: 1) LegoOS , a new distributed operating system designed for managing disaggregated resources. LegoOS splits OS functionalities into different units, each running at a hardware component and managing the component's hardware resources. LegoOS enables the disaggregation and customization of OS functionalities, a significant step towards building DC-3.0's software infrastructure. 2) LegoFPGA , a new approach of using FPGA to efficiently manage and virtualize hardware resources. LegoFPGA offers a solution to co-design application, OS, and hardware functionalities and customize them for different hardware resources and application domains, an important step towards building DC-3.0's hardware infrastructure. With LegoOS and LegoFPGA , we demonstrate that separating core OS and hardware functionalities is not only feasible but can largely improve performance per dollar over the current datacenter monolithic server model. Bio: Yiying Zhang is an assistant professor in the School of Electrical and Computer Engineering at Purdue University. Her research interests span operating systems, distributed systems, computer architecture, and datacenter networking. She also works on the intersection of systems and programming language, security, and AI/ML. She won an OSDI best paper award in 2018 and an NSF CAREER award in 2019. Yiying’s lab is among the few groups in the world now that build new OSes and full-stack, cross-layer systems. Yiying received her Ph.D. from the Department of Computer Sciences at the University of Wisconsin-Madison under the supervision of Andrea and Remzi Arpaci-Dusseau and worked as a postdoctoral scholar at the University of California, San Diego before joining Purdue.

1 0

Jennifer Gossels will present her Pre FPO "Joint Optimization for Robust Network Design and Operation" on Thursday, May 23, 2019 at 1pm in CS 401
by Nicki Mahler 17 May '19

17 May '19

Jennifer Gossels will present her Pre FPO "Joint Optimization for Robust Network Design and Operation" on Thursday, May 23, 2019 at 1pm in CS 401. The members of her committee are as follows: Jennifer Rexford (adviser), Mike Freedman (non reader), Wyatt Lloyd (non reader), Kai Li (reader), and Gagan Choudhury (AT&T; reader). Her abstract follows below. All are welcome to attend. Abstract: The Internet is an integral part of nearly all aspects of modern life, and Internet Service Provider (ISP) backbone networks play an integral part in our Internet experience. Therefore, ISPs must design their networks to limit congestion and also to limit the disruptions caused by component failures. One obvious way to ensure that there's no congestion, even when switches or links fail, is to dramatically over provision the network. If the ISP builds, e.g., two redundant links for every link it actually needs, that network will probably not get too congested or disturbed by failures. But, buying all this equipment is prohibitively costly for the ISP. In this talk we explore how the ISP can optimally balance its objective of carrying all offered traffic with minimal congestion, even in the presence of failures, with its objective of minimizing costs. The question is complicated because the ISP makes some decisions, like purchasing and placing equipment, on long timescales (months or years), and some decisions, like exactly how to route packets from a given flow, on short timescales (seconds). We classify the long timescale decisions as network design and the short timescale decisions as network operation. These decisions interact such that the ISP must consider both of them simultaneously if it wants to make a decision about either one. Accounting for failures only complicates the situation further. In the first part of this talk, we focus on the network design problem, leveraging modern optical equipment's reconfigurability to design a minimal cost network that can still carry all traffic, even if components fail. Although this part of the talk focuses on network design, we also address network operation, since the two are interdependent. Our solution allows ISPs to save up to 29% off the cost of their network infrastructure. In the second part of this talk we shift our focus to network operation. We propose a failure recovery protocol that allows for faster responses to failures than the approaches currently employed by ISPs. Our insight here is that this improved network operation approach is possible if ISPs take into account the heterogenous nature of the switches they have deployed when designing their networks. Hence, although this part of the talk focuses on network operation, it, too, addresses network design.

1 0

Yuping Luo will present his General Exam on Thursday, May 23, 2019 at 4pm in CS 301
by Nicki Mahler 16 May '19

16 May '19

Yuping Luo will present his General Exam on Thursday, May 23, 2019 at 4pm in CS 301. The members of his committee are Sanjeev Arora (adviser), Karthik Narasimhan, and Elad Hazan. 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. Abstract: In recent years, model-free reinforcement learning has demonstrated its significant success in many fields. However, these algorithms suffer from high sample complexity and require a massive number of interactions with the environment, which can be expensive in the physical world. Model-based reinforcement learning is considered to be a promising approach to reduce sample complexity. However, the theoretical understanding of such methods has been rather limited. In this talk, I'll present an algorithmic framework for designing and analyzing model-based RL algorithms with theoretical guarantees. We design a meta-algorithm with a theoretical guarantee of monotone improvement to a local maximum of the expected reward. The meta-algorithm iteratively builds a lower bound of the expected reward based on the estimated dynamical model and sample trajectories and then maximizes the lower bound jointly over the policy and the model. Instantiating our framework with simplification gives a variant of model-based RL algorithms Stochastic Lower Bounds Optimization. Experiments demonstrate that our algorithm achieves state-of-the-art performance when only one million or fewer samples are permitted on a range of continuous control benchmark tasks. [1] Thanard Kurutach, Ignasi Clavera, Yan Duan, Aviv Tamar, and Pieter Abbeel. Model-ensemble trust-region policy optimization. arXiv preprint arXiv:1802.10592, 2018. [2] Sutton, Richard S. "Dyna, an integrated architecture for learning, planning, and reacting." ACM SIGART Bulletin 2.4 (1991): 160-163. [3] Gu, Shixiang, et al. "Continuous deep q-learning with model-based acceleration." International Conference on Machine Learning. 2016. [4] Buckman, Jacob, et al. "Sample-efficient reinforcement learning with stochastic ensemble value expansion." Advances in Neural Information Processing Systems. 2018. [5] Feinberg, Vladimir, et al. "Model-based value estimation for efficient model-free reinforcement learning." arXiv preprint arXiv:1803.00101 (2018). [6] Pascanu, Razvan, et al. "Learning model-based planning from scratch." arXiv preprint arXiv:1707.06170 (2017). [7] Clavera, Ignasi, et al. "Model-based reinforcement learning via meta-policy optimization." arXiv preprint arXiv:1809.05214 (2018). [8] Chua, Kurtland, et al. "Deep reinforcement learning in a handful of trials using probabilistic dynamics models." Advances in Neural Information Processing Systems. 2018. [9] Asadi, Kavosh, Dipendra Misra, and Michael L. Littman. "Lipschitz continuity in model-based reinforcement learning." arXiv preprint arXiv:1804.07193 (2018). [10] Farahmand, Amir-massoud, Andre Barreto, and Daniel Nikovski. "Value-aware loss function for model-based reinforcement learning." Artificial Intelligence and Statistics. 2017.

1 0