Filters








61 Hits in 1.7 sec

Verifying Handcoded Probabilistic Inference Procedures [article]

Eric Atkinson, Cambridge Yang, Michael Carbin
2018 arXiv   pre-print
We have used Shuffle to develop inference algorithms for several standard probabilistic models.  ...  Researchers have recently proposed several systems that ease the process of performing Bayesian probabilistic inference.  ...  Probabilistic Inference Programming Constructing efficient inference procedures by hand, alternatively probabilistic inference programming, requires tackling several programming idioms in the domain.  ... 
arXiv:1805.01863v1 fatcat:3npadg2c3rgjpehnew6wse2qcq

Machine Learning and Model Checking Join Forces (Dagstuhl Seminar 18121)

Nils Jansen, Joost-Pieter Katoen, Pusmeet Kohli, Jan Kretinsky, Michael Wagner
2018 Dagstuhl Reports  
References 1 Verifying Handcoded Probabilistic Inference Procedures Eric Atkinson, Cambridge Yang, and Michael Carbin https://arxiv.org/abs/1805.01863 A dual approach to scalable verification of neural  ...  We have used Shuffle to develop inference algorithms for several standard probabilistic models.  ...  We consider the problem of designing machine learning models used within a larger system that must satisfy a formal specification, a step towards the goal of verified artificial intelligence (AI) [4]  ... 
doi:10.4230/dagrep.8.3.74 dblp:journals/dagstuhl-reports/JansenKKK18 fatcat:225qaztsujhgxpclahyf4wm7qe

A relational hierarchical model for decision-theoretic assistance

Sriraam Natarajan, Prasad Tadepalli, Alan Fern
2011 Knowledge and Information Systems  
In recent work, a domain-independent decision-theoretic model of assistance was proposed, where the task is to infer the user's goal and take actions that minimize the expected cost of the user's policy  ...  Systems that have been used for assistance in spreadsheets [8] and text editing [9] have used handcoded DBNs to infer about the user.  ...  We verify this empirically in our experiments.  ... 
doi:10.1007/s10115-011-0435-z fatcat:pk3olw7zvjdnlcbp2rnacmayue

Large-Scale Object Classification Using Label Relation Graphs [chapter]

Jia Deng, Nan Ding, Yangqing Jia, Andrea Frome, Kevin Murphy, Samy Bengio, Yuan Li, Hartmut Neven, Hartwig Adam
2014 Lecture Notes in Computer Science  
Next, we propose a probabilistic classification model based on HEX graphs and show that it enjoys a number of desirable properties. Finally, we evaluate our method using a large-scale benchmark.  ...  Then the online inference for each example simply follows this pre-determined sum-product sequence and thus has negligible extra overhead compared to handcoded implementations of softmax or independent  ...  For example, it is easy to verify that for pairwise mutually exclusive labels, the inference cost is O(n), the same as hand-coded softmax, due to the small state space.  ... 
doi:10.1007/978-3-319-10590-1_4 fatcat:spstchnwwvfhdhllopqglggxbq

Online probabilistic learning for fuzzy inference system

Richard J. Oentaryo, Meng Joo Er, San Linn, Xiang Li
2014 Expert systems with applications  
In light of these issues, we develop a new Sequential Probabilistic Learning for Adaptive Fuzzy Inference System (SPLAFIS) that synergizes the Bayesian Adaptive Resonance Theory (BART) and Rule-Wise Decoupled  ...  To manage the model complexity without sacrificing its predictive accuracy, SPLAFIS also includes a simple procedure to prune inconsequential rules that have little contribution over time.  ...  ., the node and link constructs) is fixed and needs to be handcoded.  ... 
doi:10.1016/j.eswa.2014.01.034 fatcat:qdnwfirw2vguvfo3xg4jj5th5e

Bayesian inference for the information gain model

Sven Stringer, Denny Borsboom, Eric-Jan Wagenmakers
2011 Behavior Research Methods  
Participants are asked which cards should be turned to verify whether or not the rule holds.  ...  In this article, we present two estimation methods to fit the information gain model: a maximum likelihood procedure (programmed in R) and a Bayesian procedure (programmed in WinBUGS).  ...  Participants must determine which cards they need to turn in order to verify whether or not this conditional rule holds.  ... 
doi:10.3758/s13428-010-0057-5 pmid:21302022 pmcid:PMC3098374 fatcat:tjvezy55lbcqveknck3wbfmcnm

Some Basic Principles of Developmental Robotics

A. Stoytchev
2009 IEEE Transactions on Autonomous Mental Development  
He can also verify this statement in the "weak" sense as he is physically capable of performing the verification procedure if necessary.  ...  If verification is not possible for some concept then the AI system should not be handcoded with that concept.  ... 
doi:10.1109/tamd.2009.2029989 fatcat:kvnz6yhsbbd7xnqaozhrv2bexy

Models of Metaphor in NLP

Ekaterina Shutova
2010 Annual Meeting of the Association for Computational Linguistics  
It originates in the work of Wilks (1978) and utilizes handcoded knowledge.  ...  Metaphor Identification Pragglejaz Procedure Pragglejaz Group (2007) proposes a metaphor identification procedure (MIP) within the frame-work of the Metaphor in Discourse project (Steen, 2007) .  ... 
dblp:conf/acl/Shutova10 fatcat:lf4mopzz25auxffsfj3mpvmhqm

BioCode: A Data-Driven Procedure to Learn the Growth of Biological Networks [article]

Emre Sefer
2021 arXiv   pre-print
We combine such instruction-wise representation with a genetic algorithm based optimization procedure to encode models for various biological networks.  ...  Well-known examples of these probabilistic models are Kronecker model, preferential attachment model, and duplication-based model.  ...  Those models describe the biological networks growth mechanistically and probabilistically.  ... 
arXiv:2108.04776v2 fatcat:uzzd4r5e6vh5xiauiprpfwqczi

From Machine Learning to Robotics: Challenges and Opportunities for Embodied Intelligence [article]

Nicholas Roy, Ingmar Posner, Tim Barfoot, Philippe Beaudoin, Yoshua Bengio, Jeannette Bohg, Oliver Brock, Isabelle Depatie, Dieter Fox, Dan Koditschek, Tomas Lozano-Perez, Vikash Mansinghka (+8 others)
2021 arXiv   pre-print
version of propositional logic), probabilistic planning languages, probabilistic programs, and temporal logics.  ...  Just as deep learning has substantially reduced the need to handcode feature functions for many problems, it is possible that the current reliance on handcoded background knowledge theories is a substantial  ... 
arXiv:2110.15245v1 fatcat:juxc4tai2jbklpul55loccnp7e

The actor's view of automated planning and acting: A position paper

Malik Ghallab, Dana Nau, Paolo Traverso
2014 Artificial Intelligence  
The information in this paper does not necessarily reflect the position or policy of the funders; no official endorsement should be inferred.  ...  In the case of nondeterministic or probabilistic models properties should be verified on sets of histories or other complex structures.  ...  In planning, some work has been done to address the problem of verifying properties of models [46] and verifying the correctness of plans.  ... 
doi:10.1016/j.artint.2013.11.002 fatcat:qrzk25vnfncwhg7kuwxghp4xom

Automatic generation of library bindings using static analysis

Tristan Ravitch, Steve Jackson, Eric Aderhold, Ben Liblit
2009 Proceedings of the 2009 ACM SIGPLAN conference on Programming language design and implementation - PLDI '09  
This work is complementary to our own: Furr and Foster verify correctness while we offer correctness by construction. Our analysis can be seen as a form of specification inference. Kremenek et al.  ...  We have verified the correctness of our results by hand in these cases.  ... 
doi:10.1145/1542476.1542516 dblp:conf/pldi/RavitchJAL09 fatcat:eypoxwd4b5ebpoy5xh6qnr4jea

Automatic generation of library bindings using static analysis

Tristan Ravitch, Steve Jackson, Eric Aderhold, Ben Liblit
2009 SIGPLAN notices  
This work is complementary to our own: Furr and Foster verify correctness while we offer correctness by construction. Our analysis can be seen as a form of specification inference. Kremenek et al.  ...  We have verified the correctness of our results by hand in these cases.  ... 
doi:10.1145/1543135.1542516 fatcat:ltb3yfcptbgdffnlocc6wmfdxe

Machine-generated algorithms, proofs and software for the batch verification of digital signature schemes

Joseph A. Akinyele, Matthew Green, Susan Hohenberger, Matthew W. Pagano
2012 Proceedings of the 2012 ACM conference on Computer and communications security - CCS '12  
Boneh, Lynn and Shacham provided a single-signer batch verifier for BLS signatures [13] .  ...  Pairing-based signatures can be very short, but are often costly to verify. Fortunately, they also tend to have efficient batch verification algorithms.  ...  We obtain this information by applying known rules to infer type.  ... 
doi:10.1145/2382196.2382248 dblp:conf/ccs/Akinyele0HP12 fatcat:f5fqwtbxrbgmth67hltvz4pbiq

Machine-generated algorithms, proofs and software for the batch verification of digital signature schemes

Joseph A. Akinyele, Matthew Green, Susan Hohenberger, Matthew Pagano
2014 Journal of Computer Security  
Boneh, Lynn and Shacham provided a single-signer batch verifier for BLS signatures [13] .  ...  Pairing-based signatures can be very short, but are often costly to verify. Fortunately, they also tend to have efficient batch verification algorithms.  ...  We obtain this information by applying known rules to infer type.  ... 
doi:10.3233/jcs-140507 fatcat:lrmq3wnlhzgxhew6yk3akzn6yy
« Previous Showing results 1 — 15 out of 61 results