Filters








228 Hits in 3.2 sec

Gaining degrees of freedom in subsymbolic learning

B. Apolloni, D. Malchiodi
2001 Theoretical Computer Science  
Thus, for a given concept class and a given sample size, we discuss the e ciency of subsymbolical learning algorithms in terms of degrees of freedom of the computed statistic.  ...  In this framework we give new foundations to the notion of degrees of freedom of a statistic and relate it to the complexity of a concept class.  ...  In these pages, we try to narrow the gap between symbolic theory and subsymbolic practice, specialising the degrees of freedom of a labelled sample in some widespread learning instances.  ... 
doi:10.1016/s0304-3975(99)00289-3 fatcat:sbskdexgsrauvh47toiclcu5d4

Page 7542 of Mathematical Reviews Vol. , Issue 2002J [page]

2002 Mathematical Reviews  
(I-MILAN; Milan) Gaining degrees of freedom in subsymbolic learning. (English summary ) Theoret. Comput. Sci. 255 (2001), no. 1-2, 295-321.  ...  Thus, for a given concept class and a given sample size, we discuss the ef- ficiency of subsymbolical learning algorithms in terms of degrees of freedom of the computed statistic.  ... 

Learning accurate, compact, and interpretable tree annotation

Slav Petrov, Leon Barrett, Romain Thibaux, Dan Klein
2006 Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the ACL - ACL '06  
Starting with a simple Xbar grammar, we learn a new grammar whose nonterminals are subsymbols of the original nonterminals.  ...  Our grammars automatically learn the kinds of linguistic distinctions exhibited in previous work on manual tree annotation.  ...  However, in general, any automatic induction system is in danger of being entirely uninterpretable. In this section, we examine the learned grammars, discussing what is learned.  ... 
doi:10.3115/1220175.1220230 dblp:conf/acl/PetrovBTK06 fatcat:gslixhjfhbfglkkfxbemefr6zq

The Infinite PCFG Using Hierarchical Dirichlet Processes

Percy Liang, Slav Petrov, Michael I. Jordan, Dan Klein
2007 Conference on Empirical Methods in Natural Language Processing  
We also show that our techniques can be applied to full-scale parsing applications by demonstrating its effectiveness in learning state-split grammars.  ...  Our HDP-PCFG model allows the complexity of the grammar to grow as more training data is available.  ...  This lack of normalization gives an extra degree of freedom not present in maximum likelihood estimation: it creates a global preference for left-hand sides that have larger total counts.  ... 
dblp:conf/emnlp/LiangPJK07 fatcat:2ok3gdhvbjfqne72bzrpr7dwu4

Moral agency without responsibility? Analysis of three ethical models of human-computer interaction in times of artificial intelligence (AI)

Alexis Fritz, Wiebke Brandt, Henner Gimpel, Sarah Bayer
2020 De Ethica  
Initially, the division between symbolic and sub-symbolic AI, the black box character of (deep) machine learning, and the complex relationship network in the provision and application of machine learning  ...  It is only in this way – so their principal argument goes – that the effects of technological components in a complex human-computer interaction can be understood sufficiently in phenomenological-descriptive  ...  Deep learning is a form of machine learning that has gained popularity in recent years due to advances in (big) data availability, (cloud-based) massive computing power, algorithms, and openly available  ... 
doi:10.3384/de-ethica.2001-8819.20613 fatcat:bam5fn2xyjeqxg54ltouhv5dfa

A PSYCHOGENETIC ALGORITHM FOR BEHAVIORAL SEQUENCE LEARNING

VITTORIO MANIEZZO, MATTEO ROFFILLI
2007 International journal on artificial intelligence tools  
learning or, more in general, of Reinforced Learning (RL).  ...  Further, we discussed the possible parallels between our model and subsymbolic machine learning and neuroscience.  ...  In order to better capture the laws inside the input data, the network construction proceeds by adding hidden layers (degrees of freedom) to the basic perceptron architecture (linear kernel function).  ... 
doi:10.1142/s021821300700328x fatcat:l6ex6qif2rdvzagik5miz22fle

Author index volume 255 (2001)

2001 Theoretical Computer Science  
Malchiodi, Gaining degrees of freedom in subsymbolic learning (1}2) 295}321 Arikati, S.R., A. Dessmark, A. Lingas and M.V.  ...  Chung, Improved fault-tolerant sorting algorithm in hypercubes (Note) (1}2) 649}658 Chung, K.-L., see Y.-W. Chen (1}2) 649}658 Csuhaj-VarjuH , E. and G.  ... 
doi:10.1016/s0304-3975(01)00041-x fatcat:g2gqkmgyu5hxtp6i7jwlixjzwa

Argument Schemes for AI Ethics Education

Nancy L. Green, L. Joshua Crotts
2020 Computational Models of Argument  
It then describes use of the schemes in an argument diagramming tool and results of a formative evaluation.  ...  Joshua Crotts programmed AIED in summer 2019 with support from a University of North Carolina Greensboro Faculty First Award.  ...  A is ethically acceptable in S to some degree. 2. B is ethically acceptable in S to some degree. 3.  ... 
dblp:conf/comma/GreenC20 fatcat:fug6dqu2crglzcx256yt6gc77i

The semiotic life cycle and The Symbolic Species

Tyler James Bennett
2015 Sign Systems Studies  
Th e problem is not that Peirce should not be used in this way. In fact Deacon's book is a singular achievement in the application of Peirce.  ...  In fact Deacon's claim about the possible disadvantages of symbol use can be reinforced with a closer look at the mature, turn-of-the-century Peircean sign model.  ...  Th e degree of freedom of agency symbolic legisigns actually possess seems intentionally exaggerated by Nöth, but towards the end of the article even he pulls back enough to say that, whatever agency symbolic  ... 
doi:10.12697/sss.2015.43.4.05 fatcat:7jc67wyy4rh33prhftyfcifake

Combining Representation Learning with Logic for Language Processing [article]

Tim Rocktäschel
2017 arXiv   pre-print
However, in many cases representation learning requires a large amount of annotated training data to generalize well to unseen data.  ...  The current state-of-the-art in many natural language processing and automated knowledge base completion tasks is held by representation learning methods which learn distributed vector representations  ...  Note that so far we have not gained anything over matrix factorization as explained in Section 2.3.1.  ... 
arXiv:1712.09687v1 fatcat:mfpom6sfkfaj3a3znsiaxjrrme

Operational Performance of an Automatic Preliminary Spectral Rule-Based Decision-Tree Classifier of Spaceborne Very High Resolution Optical Images

Andrea Baraldi, Tom Wassenaar, Simon Kay
2010 IEEE Transactions on Geoscience and Remote Sensing  
Given the ISRC "full degree" of automation, which cannot be surpassed, and ISRC computation time, which is near real time, this paper provides a quantitative assessment of ISRC accuracy and robustness  ...  In the last 20 years, the number of spaceborne very high resolution (VHR) optical imaging sensors and the use of satellite VHR optical images have continued to increase both in terms of quantity and quality  ...  ACKNOWLEDGMENT The authors would like to thank the Editor-in-Chief, the Associate Editor, and anonymous reviewers for their patience  ... 
doi:10.1109/tgrs.2010.2046741 fatcat:vmohwianbbcqzdnwm3afuyt7eu

Brittle Opacity: Ambiguities of the Creative AI

Dejan Grba
2021 Zenodo  
This paper outlines the ambiguities which influence AI science, manifest in the production of AI artists, and shape the representation of creative AI in the media and in popular culture.  ...  creative investigation of their tools, and in more nuanced scrutiny of their work.  ...  layer, the degree of change in each weight during learning, and many other (Mitchell 2019a, 82) . t Randomized Living qualifies as a strong artwork of cybernetic-existentialismthe art of conceiving a  ... 
doi:10.5281/zenodo.5831884 fatcat:qqdfj4pyqnaqhavfr5ns5tvfmi

Integrating Intrinsic and Extrinsic Explainability: The Relevance of Understanding Neural Networks for Human-Robot Interaction [article]

Tom Weber, Stefan Wermter
2020 arXiv   pre-print
Explainable artificial intelligence (XAI) can help foster trust in and acceptance of intelligent and autonomous systems.  ...  However, a lot of these approaches are not applicable to humanoid robots. Therefore, in this position paper, current problems with adapting XAI methods to explainable neurorobotics are described.  ...  Lastly, two degrees of freedom (DoF) for yaw and pitch head movements and at least eight DoF per arm, depending on the exact hand model, allow to convey information via complex gestures.  ... 
arXiv:2010.04602v1 fatcat:k2xqu4heyjalxltcseln5cy554

Nonclassical connectionism should enter the decathlon

Francisco Calvo Garzón
2003 Behavioral and Brain Sciences  
In contrast, ACT-R includes both symbolic and subsymbolic components. The strengths of the ACT-R theory derive from its tight integration of the symbolic component with the subsymbolic component.  ...  The strengths of classical connectionism on this test derive from its intense effort in addressing empirical phenomena in such domains as language and cognitive development.  ...  ACKNOWLEDGMENTS ACKNOWLEDGMENT This work was supported in part by ARI contract DASW01-00 -K-0012.  ... 
doi:10.1017/s0140525x03240139 fatcat:qzqkqm6ejvelzostaoc4lfp6ni

Active exploration of joint dependency structures

Johannes Kulick, Stefan Otte, Marc Toussaint
2015 2015 IEEE International Conference on Robotics and Automation (ICRA)  
Being able to manipulate degrees of freedom of the environment, such as doors or drawers, is a requirement for most tasks a robot is supposed to perform.  ...  Often these external degrees of freedom depend on other ones, e. g., a drawer can only be opened if the lock is not locking the joint.  ...  ACKNOWLEDGMENTS This work is partially funded by the DFG (German Science Foundation) within Priority Programme 1527, "Autonomous Learning".  ... 
doi:10.1109/icra.2015.7139549 dblp:conf/icra/KulickOT15 fatcat:uygtrnx7wvamnbtfc2jw5bgcke
« Previous Showing results 1 — 15 out of 228 results