Filters








4,604 Hits in 1.9 sec

Empirically Verifying Hypotheses Using Reinforcement Learning [article]

Kenneth Marino, Rob Fergus, Arthur Szlam, Abhinav Gupta
2020 arXiv   pre-print
This paper formulates hypothesis verification as an RL problem. Specifically, we aim to build an agent that, given a hypothesis about the dynamics of the world, can take actions to generate observations which can help predict whether the hypothesis is true or false. Existing RL algorithms fail to solve this task, even for simple environments. In order to train the agents, we exploit the underlying structure of many hypotheses, factorizing them as pre-condition, action sequence, post-condition
more » ... iplets. By leveraging this structure we show that RL agents are able to succeed at the task. Furthermore, subsequent fine-tuning of the policies allows the agent to correctly verify hypotheses not amenable to the above factorization.
arXiv:2006.15762v1 fatcat:n2jiebuhrzadnpmlneotsukzsq

The Pose Knows: Video Forecasting by Generating Pose Futures [article]

Jacob Walker, Kenneth Marino, Abhinav Gupta, Martial Hebert
2017 arXiv   pre-print
Current approaches in video forecasting attempt to generate videos directly in pixel space using Generative Adversarial Networks (GANs) or Variational Autoencoders (VAEs). However, since these approaches try to model all the structure and scene dynamics at once, in unconstrained settings they often generate uninterpretable results. Our insight is to model the forecasting problem at a higher level of abstraction. Specifically, we exploit human pose detectors as a free source of supervision and
more » ... eak the video forecasting problem into two discrete steps. First we explicitly model the high level structure of active objects in the scene---humans---and use a VAE to model the possible future movements of humans in the pose space. We then use the future poses generated as conditional information to a GAN to predict the future frames of the video in pixel space. By using the structured space of pose as an intermediate representation, we sidestep the problems that GANs have in generating video pixels directly. We show through quantitative and qualitative evaluation that our method outperforms state-of-the-art methods for video prediction.
arXiv:1705.00053v1 fatcat:qao4ijregvh5xhsyvxosv4boj4

Nuevo esquema estratigráfico para los depósitos marinos mio-pliocenos del área de Navidad (33º00'-34º30'S), Chile central

Alfonso Encinas, Jacobus P Le Roux, Luis A Buatois, Sven N Nielsen, Kenneth L Finger, Elizabeth Fourtanier, Alain Lavenu
2006 Revista Geológica de Chile  
Darwin (1846) describió por primera vez los estratos marinos neógenos del área de Navidad.  ...  INTRODUCCIÓN Los depósitos marinos neógenos que afloran en el área de Navidad (~33°00'-34°30'S) ( Fig. 1 ) fueron estudiados por primera vez por Darwin (1846) quien visitó dicha zona durante su viaje  ... 
doi:10.4067/s0716-02082006000200002 fatcat:xmxxmo4oofeqpnoarq2t3fqjce

Piezoelectricity in the human pineal gland

Sidney B. Lang, Andrew A. Marino, Garry Berkovic, Marjorie Fowler, Kenneth D. Abreo
1996 Bioelectrochemistry and bioenergetics (Print)  
Melatonin secretion by the pineal gland has been reported to be affected by exposure to electromagnetic fields (EMFs). In an initial investigation to determine if calcifications commonly found in the pineal gland could respond to EMFs by a transducer mechanism, studies were conducted to ascertain if pineal tissues were piezoelectric. Second harmonic generation (SHG) measurements showed that pineal tissues contained noncentrosymmetric crystals, thus proving the presence of piezoelectricity. Both
more » ... mulberry-like and faceted crystalline calcifications were observed by scanning electron microscopy (SEM). Some of the calcifications had compositions similar to that of hydroxyapatite; others contained a high concentration of aluminum.
doi:10.1016/s0302-4598(96)05147-1 fatcat:amsgper7hfdnfghrlxmcvf57b4

The More You Know: Using Knowledge Graphs for Image Classification [article]

Kenneth Marino, Ruslan Salakhutdinov, Abhinav Gupta
2017 arXiv   pre-print
One characteristic that sets humans apart from modern learning-based computer vision algorithms is the ability to acquire knowledge about the world and use that knowledge to reason about the visual world. Humans can learn about the characteristics of objects and the relationships that occur between them to learn a large variety of visual concepts, often with few examples. This paper investigates the use of structured prior knowledge in the form of knowledge graphs and shows that using this
more » ... edge improves performance on image classification. We build on recent work on end-to-end learning on graphs, introducing the Graph Search Neural Network as a way of efficiently incorporating large knowledge graphs into a vision classification pipeline. We show in a number of experiments that our method outperforms standard neural network baselines for multi-label classification.
arXiv:1612.04844v2 fatcat:r536v42zfbccpfkkptpzolcpdq

Psychological Antecedents of Work Role Perceptions: Business Founders and Managers

Richard Castaldi, Kenneth Marino
1970 Journal of business strategies  
The work roles were determined by correlating the response of Spring 1990 Castaldi & Marino: Perceptions 57 pairs of 93 subjects.  ... 
doi:10.54155/jbs.7.1.54-65 fatcat:vxyahhhg4vgo7a76hh4qfsnfpu

OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge [article]

Kenneth Marino, Mohammad Rastegari, Ali Farhadi, Roozbeh Mottaghi
2019 arXiv   pre-print
Visual Question Answering (VQA) in its ideal form lets us study reasoning in the joint space of vision and language and serves as a proxy for the AI task of scene understanding. However, most VQA benchmarks to date are focused on questions such as simple counting, visual attributes, and object detection that do not require reasoning or knowledge beyond what is in the image. In this paper, we address the task of knowledge-based visual question answering and provide a benchmark, called OK-VQA,
more » ... re the image content is not sufficient to answer the questions, encouraging methods that rely on external knowledge resources. Our new dataset includes more than 14,000 questions that require external knowledge to answer. We show that the performance of the state-of-the-art VQA models degrades drastically in this new setting. Our analysis shows that our knowledge-based VQA task is diverse, difficult, and large compared to previous knowledge-based VQA datasets. We hope that this dataset enables researchers to open up new avenues for research in this domain. See http://okvqa.allenai.org to download and browse the dataset.
arXiv:1906.00067v2 fatcat:h2yumclhjbdtxdge4pop5lm62m

Perspective on Cerebral Autoregulation Monitoring in Neonatal Cardiac Surgery Requiring Cardiopulmonary Bypass

Jared M. Spilka, Conor P. O'Halloran, Bradley S. Marino, Kenneth M. Brady
2021 Frontiers in Neurology  
Copyright © 2021 Spilka, O'Halloran, Marino and Brady. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY).  ... 
doi:10.3389/fneur.2021.740185 pmid:34675872 pmcid:PMC8523884 fatcat:olbwnnj5tbgdxdutj3vhicsmkq

Same Object, Different Grasps: Data and Semantic Knowledge for Task-Oriented Grasping [article]

Adithyavairavan Murali, Weiyu Liu, Kenneth Marino, Sonia Chernova, Abhinav Gupta
2020 arXiv   pre-print
Despite the enormous progress and generalization in robotic grasping in recent years, existing methods have yet to scale and generalize task-oriented grasping to the same extent. This is largely due to the scale of the datasets both in terms of the number of objects and tasks studied. We address these concerns with the TaskGrasp dataset which is more diverse both in terms of objects and tasks, and an order of magnitude larger than previous datasets. The dataset contains 250K task-oriented
more » ... for 56 tasks and 191 objects along with their RGB-D information. We take advantage of this new breadth and diversity in the data and present the GCNGrasp framework which uses the semantic knowledge of objects and tasks encoded in a knowledge graph to generalize to new object instances, classes and even new tasks. Our framework shows a significant improvement of around 12% on held-out settings compared to baseline methods which do not use semantics. We demonstrate that our dataset and model are applicable for the real world by executing task-oriented grasps on a real robot on unknown objects. Code, data and supplementary video could be found at https://sites.google.com/view/taskgrasp
arXiv:2011.06431v2 fatcat:52zg6zdzxrapjbxod5j6tlfkbm

Implementing telestroke to reduce the burden of stroke in Louisiana

Praphul Joshi, Marisa Marino, Alok Bhoi, Kenneth Gaines, Elizabeth Allen, Julia Mora
2013 Journal of Cardiovascular Disease Research  
Cardiovascular diseases (heart diseases and stroke) are the leading cause of death in the United States. People living in rural areas have disproportionately high mortality rates due to stroke mainly due to lack of specialty services. Timely intervention is a critical factor in determining the prognosis for acute stroke and thousands of lives can be saved by recognizing/treating a stroke patient within an hour. This paper deals with telestroke, a unique intervention implemented in Louisiana
more » ... incorporated latest technology to integrate with advanced clinical protocols in treating acute stroke. People living in remote and rural areas of Louisiana were greatly benefited from this initiative and several lives saved as a result of timely intervention. Expanding these initiatives to larger populations and other states will greatly reduce the stroke mortality.
doi:10.1016/j.jcdr.2013.02.015 pmid:24023480 pmcid:PMC3758068 fatcat:6kd6vssncfg4binndjsxs7arny

OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge

Kenneth Marino, Mohammad Rastegari, Ali Farhadi, Roozbeh Mottaghi
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Visual Question Answering (VQA) in its ideal form lets us study reasoning in the joint space of vision and language and serves as a proxy for the AI task of scene understanding. However, most VQA benchmarks to date are focused on questions such as simple counting, visual attributes, and object detection that do not require reasoning or knowledge beyond what is in the image. In this paper, we address the task of knowledge-based visual question answering and provide a benchmark, called OK-VQA,
more » ... re the image content is not sufficient to answer the questions, encouraging methods that rely on external knowledge resources. Our new dataset includes more than 14,000 questions that require external knowledge to answer. We show that the performance of the state-of-the-art VQA models degrades drastically in this new setting. Our analysis shows that our knowledge-based VQA task is diverse, difficult, and large compared to previous knowledge-based VQA datasets. We hope that this dataset enables researchers to open up new avenues for research in this domain.
doi:10.1109/cvpr.2019.00331 dblp:conf/cvpr/MarinoRFM19 fatcat:5c4z34e7afafzov3irhjvsayia

A third fibronectin type III domain in the extracellular region of the insulin receptor family

Cristina Marino-Buslje, Kenji Mizuguchi, Kenneth Siddle, Tom L. Blundell
1998 FEBS Letters  
Acknowledgements: Cristina Marino-Buslje is a European Community fellow. Kenji Mizuguchi thanks HFSP and SmithKline Beecham for ¢nancial support.  ... 
doi:10.1016/s0014-5793(98)01509-9 pmid:9883910 fatcat:4b3m6tdbfrhtxea4mdy2t5s2ja

Metabolism of a nitrogen-enriched coastal marine lagoon during the summertime

Robert W. Howarth, Melanie Hayn, Roxanne M. Marino, Neil Ganju, Kenneth Foreman, Karen McGlathery, Anne E. Giblin, Peter Berg, Jeffrey D. Walker
2013 Biogeochemistry  
Wind velocity was normalized to a 10-m height as in Marino and Howarth (1993) .  ...  Open circles represent observations made with a floating dome technique in the Hudson River estuary (Marino and Howarth 1993) .  ... 
doi:10.1007/s10533-013-9901-x fatcat:f6ifxpyzzbde5iwo4z3zbkq4f4

Diagnostic assays for active infection with human herpesvirus 6 (HHV-6)

Mary T. Caserta, Caroline Breese Hall, Kenneth Schnabel, Geraldine Lofthus, Andrea Marino, Lynne Shelley, Christina Yoo, Jennifer Carnahan, Linda Anderson, Hongyue Wang
2010 Journal of Clinical Virology  
Background-Human herpesvirus 6 (HHV-6) causes ubiquitous infection in early childhood with lifelong latency or persistence. Reactivation of HHV-6 has been associated with multiple diseases including encephalitis. Chromosomal integration of HHV-6 also occurs. Previous studies have suggested that the detection of HHV-6 DNA in plasma is an accurate marker of active viral replication. Objective-We sought to determine whether PCR assays on plasma could correctly differentiate between primary HHV-6
more » ... fection, chromosomal integration of HHV-6 and latent HHV-6 infection. Study Design-We performed qualitative PCR, real-time quantitative PCR (RQ-PCR), and reverse-transcriptase PCR (RT-PCR) assays on samples of peripheral and cord blood mononuclear cells, as well as plasma, from groups of subjects with well defined HHV-6 infection, including subjects with chromosomally integrated HHV-6. Results and Conclusions-The detection of HHV-6 DNA in plasma was 92% sensitive compared to viral isolation for the identification of primary infection with HHV-6. All plasma samples from infants with chromosomally integrated HHV-6 had HHV-6 DNA detectable in plasma while only 5.6% were positive by RT-PCR. The specificity of plasma PCR for active replication of HHV-6 was 84% compared to viral culture while the specificity of RT-PCR was 98%. Our results demonstrate that qualitative or quantitative PCR of plasma is insufficient to distinguish between active viral replication and chromosomal integration with HHV-6. We found
doi:10.1016/j.jcv.2010.02.007 pmid:20211581 pmcid:PMC2855742 fatcat:uhybyftbo5eyrdlc6rl5nsf4ru

Evaluation of electronic biopsy for clinical diagnosis in virtual colonoscopy

Joseph Marino, Wei Du, Matthew Barish, Ellen Li, Wei Zhu, Arie Kaufman, Kenneth H. Wong, David R. Holmes III
2011 Medical Imaging 2011: Visualization, Image-Guided Procedures, and Modeling  
Virtual colonoscopy provides techniques not available in optical colonoscopy, an exciting one being the ability to perform an electronic biopsy. An electronic biopsy image is created using ray-casting volume rendering of the CT data with a translucent transfer function mapping higher densities to red and lower densities to blue. The resulting image allows the physician to gain insight into the internal structure of polyps. Benign tissue and adenomas can be differentiated; the former will appear
more » ... as homogeneously blue and the latter as irregular red structures. Although this technique is now common, is included with clinical systems, and has been used successfully for computer aided detection, there has so far been no study to evaluate the effectiveness of a physician using electronic biopsy in determining the pathological state of a polyp. We present here such a study, wherein an experienced radiologist ranked polyps based on electronic biopsy alone per scan (supine and prone), as well as both combined. Our results show a correct identification 77% of the time using prone or supine images alone, and 80% accuracy using both. Using ROC analysis based on this study with one reader and a modest sample size, the combined score is not significantly higher than using a single electronic biopsy image alone. However, our analysis indicates a trend of superiority for the combined ranking that deserves a follow-up confirmatory study with a larger sample and more readers. This study yields hope that an improved electronic biopsy technique could become a primary clinical diagnosis method.
doi:10.1117/12.878295 dblp:conf/miigp/MarinoDBLZK11 fatcat:kk6gpd7jz5dizbxze4sbbftgt4
« Previous Showing results 1 — 15 out of 4,604 results