Filters








59,356 Hits in 7.2 sec

Jointly Learning "What" and "How" from Instructions and Goal-States

Dzmitry Bahdanau, Felix Hill, Jan Leike, Edward Hughes, Pushmeet Kohli, Edward Grefenstette
2018 International Conference on Learning Representations  
For these scenarios, we introduce a method which requires only pairs of instructions and examples of goal states, from which we can jointly learn a model of the instruction-conditional reward and a policy  ...  We furthermore evaluate the generalization of our approach to unseen instructions, and to scenarios where environment dynamics change outside of training, requiring fine-tuning of the policy "in the wild  ...  Given a dataset of instructions and respective goal-state examples provided by an expert, we train a discriminator network and a policy network that focus on the "what to do" and "how to do it" aspects  ... 
dblp:conf/iclr/BahdanauHLHKG18 fatcat:liku7o5dnfcgpk6nlk7xgf43oi

Learning to Jointly Predict Ellipsis and Comparison Structures

Omid Bakhshandeh, Alexis Cornelia Wellwood, James Allen
2016 Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning  
In this paper, we introduce a novel framework for jointly capturing the semantic structure of comparison and ellipsis constructions.  ...  Comparison plays a crucial role in shaping objective and subjective opinion and measurement in natural language, and is often expressed in complex constructions including ellipsis.  ...  Acknowledgment We thank the anonymous reviewers for their invaluable comments and Brian Rinehart and other annotators for their great work on the annotations.  ... 
doi:10.18653/v1/k16-1007 dblp:conf/conll/BakhshandehWA16 fatcat:54oiz73gsfbgzjbbaezmwy6kca

Jointly Learning Grounded Task Structures from Language Instruction and Visual Demonstration

Changsong Liu, Shaohua Yang, Sari Saba-Sadiya, Nishant Shukla, Yunzhong He, Song-chun Zhu, Joyce Chai
2016 Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing  
To enable language-based communication and collaboration with cognitive robots, this paper presents an approach where an agent can learn task models jointly from language instruction and visual demonstration  ...  The learned AoG captures a hierarchical task structure where linguistic labels (for language communication) are grounded to corresponding state changes from the physical environment (for perception and  ...  This work was supported in part by N66001-15-C-4035 from the DARPA SIMPLEX program, and IIS-1208390 and IIS-1617682 from the National Science Foundation.  ... 
doi:10.18653/v1/d16-1155 dblp:conf/emnlp/LiuYSSHZC16 fatcat:qwboe4kby5gwliwyhbp4yjrssi

Jointly Learning Energy Expenditures and Activities Using Egocentric Multimodal Signals

Katsuyuki Nakamura, Serena Yeung, Alexandre Alahi, Li Fei-Fei
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Physiological signals such as heart rate can provide valuable information about an individual's state and activity.  ...  In this work, we propose a model for reasoning on multimodal data to jointly predict activities and energy expenditures.  ...  Acknowledgements Serena Yeung, Alexandre Alahi and Li Fei-Fei are partially supported by a grant from Intel.  ... 
doi:10.1109/cvpr.2017.721 dblp:conf/cvpr/NakamuraYAF17 fatcat:zlqt4iaj2nc2jivjda2nafmkkm

Oscillatory and aperiodic neural activity jointly predict grammar learning [article]

Zachariah R. Cross, Andrew W. Corcoran, Matthias Schlesewsky, Mark. J. Kohler, Ina Bornkessel-Schlesewsky
2020 bioRxiv   pre-print
Using spectral decomposition and power-law exponent estimation of human EEG data, we show for the first time that 1/f and oscillatory activity jointly influence the learning of different word order rules  ...  Further, recent work has shown that aperiodic (non-oscillatory) 1/f activity is functionally and behaviourally relevant, yet its interaction with oscillatory activity during complex rule learning remains  ...  Estimates derived from pre-and post-session eyes-closed resting states were then averaged for each participant using meanIAF.m.  ... 
doi:10.1101/2020.03.10.984971 fatcat:tlisn7shubbzhigw6b24vg3kxe

Cue integration in spatial search for jointly learned landmarks but not for separately learned landmarks

Yu Du, Neil McMillan, Christopher R. Madan, Marcia L. Spetch, Weimin Mou
2017 Journal of Experimental Psychology. Learning, Memory and Cognition  
The authors investigated how humans use multiple landmarks to locate a goal. Participants searched for a hidden goal location along a line between 2 distinct landmarks on a computer screen.  ...  In Experiment 1, some baseline trials provided both landmarks, and some provided only 1 landmark. On probe trials, both landmarks were shifted apart relative to the previously learned goal location.  ...  ., identical to test trials) and instructions at the beginning of this phase stated that some trials would not give feedback.  ... 
doi:10.1037/xlm0000416 pmid:28504533 fatcat:nmj3qwytsndndf6blpevdpoh5i

Learning from Explicit and Implicit Supervision Jointly For Algebra Word Problems

Shyam Upadhyay, Ming-Wei Chang, Kai-Wei Chang, Wen-tau Yih
2016 Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing  
To leverage the mined dataset, we propose a novel structured-output learning algorithm that aims to learn from both explicit (e.g., equations) and implicit (e.g., solutions) supervision signals jointly  ...  Existing state-of-the-art approaches mainly rely on learning from human annotated equations.  ...  We argue that the effective strategy is to learn from both explicit and implicit supervision signals jointly.  ... 
doi:10.18653/v1/d16-1029 dblp:conf/emnlp/UpadhyayCCY16 fatcat:l6k4uxyki5amvceffgtuvrp274

Synthetic Data – what, why and how? [article]

James Jordon, Lukasz Szpruch, Florimond Houssiau, Mirko Bottarelli, Giovanni Cherubin, Carsten Maple, Samuel N. Cohen, Adrian Weller
2022 arXiv   pre-print
This explainer document aims to provide an overview of the current state of the rapidly expanding work on synthetic data technologies, with a particular focus on privacy.  ...  We do believe that synthetic data is a very useful tool, and our hope is that this report highlights that, while drawing attention to nuances that can easily be overlooked in its deployment.  ...  Acknowledgements We would like to thank Accenture, Hazy, HSBC, MOSTLY AI, and the Office for National Statistics for their participation in our interviews.  ... 
arXiv:2205.03257v1 fatcat:2sdjqyfdivb4hnxn3jr2j5e6q4

Design Science: Why, What and How

Panos Y. Papalambros
2015 Design Science  
In the early years when design research was starting to define the field of design and topics of research, the newly established design community shared a common goal that was mainly related to the complexity  ...  of design and the need to understand design and to support the design process.  ...  The most common concept mentioned, as expected, is design science: what it is and what its goals and roles are.  ... 
doi:10.1017/dsj.2015.1 fatcat:eyat6vzngndozfhcjbaliab3p4

Military Professionalization Programs in Kazakhstan and the United States: How to Implement and What Will We Gain?

Sebastian Engels
2017 Connections The Quarterly Journal  
The U.S. should remain committed to Central Asian security cooperation, but must carefully evaluate each program for merit and value added to U.S. security goals in the region.  ...  Training programs that professionalize the Kazakh military can offer a cost-effective way for the United States to further a lasting partnership with Central Asia's most stable country.  ...  The right people and organizations from both sides need to cooperate, work together to build a fruitful relationship, and continually and jointly assess their endeavors.  ... 
doi:10.11610/connections.16.2.06 fatcat:reevthf3c5d6zngxx3wkdr2uwe

Jointly Learning to Parse and Perceive: Connecting Natural Language to the Physical World

Jayant Krishnamurthy, Thomas Kollar
2013 Transactions of the Association for Computational Linguistics  
LSP learns physical representations for both categorical ("blue," "mug") and relational ("on") language, and also learns to compose these representations to produce the referents of entire statements.  ...  We perform experiments on two applications: scene understanding and geographical question answering.  ...  Acknowledgments This research has been supported in part by DARPA under award FA8750-13-2-0005, and in part by a gift from Google.  ... 
doi:10.1162/tacl_a_00220 fatcat:45hhhzzqhbbj5p7u6skp53xspa

Investigating how and what prospective teachers learn through microteaching lesson study

Maria Lorelei Fernández
2010 Teaching and Teacher Education : An International Journal of Research and Studies  
Active learning involving meaningful discussion, planning, and practice, support from a knowledgeable advisor, collaborative deliberation-in-process, and opportunity to trial, analyse and revise were aspects  ...  The pre-and post-lesson plans demonstrated growth in participants' knowledge of teaching aligned with an overarching student-learning process goal (i.e., mathematics reasoning).  ...  Developing knowledge and skills of content-related instructional approaches, understanding of how students learn content through learning processes, and developing understanding of the content are all  ... 
doi:10.1016/j.tate.2009.09.012 fatcat:2msttjlkobgptj5kygfv2sy22a

What is Scientific Thinking and How Does it Develop? [chapter]

Deanna Kuhn
2010 The Wiley-Blackwell Handbook of Childhood Cognitive Development  
If we can clearly identify what the cognitive skills are and how they develop, we are in the best position to learn how to promote understanding of their value.  ...  A group of people may rely jointly on scientific thinking in pursuing their goals.  ... 
doi:10.1002/9781444325485.ch19 fatcat:sdk7c46oyfgxdk3hgccbnrnh7m

Supplementing Coursebooks--How, Why, and to What Effect: A Case Study

John Harper
2019 Journal of English Language Teaching and Linguistics  
While findings also indicated that learners generally appreciated the instructor's use of supplementary materials, significant mismatches between instructor goals and learner interpretations suggested  ...  Findings indicated that the instructor provided a variety of types of supplementary materials in order to cater to different learner needs (specifically, affective needs, language-learning needs, course-content  ...  In discussing the value of intercultural communication-based supplements, Charlie stated the following: -What we learn is the different culture.  ... 
doi:10.21462/jeltl.v4i2.283 fatcat:z6ddkd4zmbf4bothgk64rnn6ye

Teaching how to learn with a wiki in primary education: What classroom interaction can tell us

M. Pifarré, Li Li
2012 Learning, Culture and Social Interaction  
Understanding how wikis are used to support collaborative learning is an important concern for researchers and teachers.  ...  Through studying interaction between the teacher and students, our findings identify ways in which the teacher prompts collaborative learning but also shed light on the difficulties for the teacher in  ...  The authors would like to thank the teacher and the pupils for their participation in the study reported in this paper. References  ... 
doi:10.1016/j.lcsi.2012.05.004 fatcat:jbkh5xoftbhalgwqvaiaicwf5y
« Previous Showing results 1 — 15 out of 59,356 results