Filters








780 Hits in 4.7 sec

Steering model-based oracles to admit real program behaviors

Gregory Gay, Sanjai Rayadurgam, Mats P. E. Heimdahl
2014 Companion Proceedings of the 36th International Conference on Software Engineering - ICSE Companion 2014  
We therefore propose steering the model when used as an oracle, to admit an expanded set of behaviors when judging the SUT's adherence to its requirements.  ...  Early results show that steering significantly increases SUT-oracle conformance with minimal masking of real faults and, thus, has significant potential for reducing development costs.  ...  We take inspiration from program steering-however, instead of steering the SUT, we steer the oracle to see if the model is capable of matching the SUT's behavior.  ... 
doi:10.1145/2591062.2591120 dblp:conf/icse/GayRH14 fatcat:f5adamdcyncalahwdz756j465q

Automated Steering of Model-Based Test Oracles to Admit Real Program Behaviors

Gregory Gay, Sanjai Rayadurgam, Mats P. E. Heimdahl
2017 IEEE Transactions on Software Engineering  
Automated Steering of Model-Based Test Oracles to Admit Real Program Behaviors.  ...  Steering Model-Based Oracles to Admit Real Program Behaviors. In Proceedings of the 36th International Conference on Software Engineering (ICSE Companion 2014).  ...  In our work, we have used behavioral models as oracle information. Other oracles may use a set of assertions, or expected values for certain variables.  ... 
doi:10.1109/tse.2016.2615311 fatcat:iholcljqenbcnjs3lklzdpg6aq

A Formal and Quantifiable Log Analysis Framework for Test Driving of Autonomous Vehicles

Kyungbok Sung, Kyoung-Wook Min, Jeongdan Choi, Byung-Cheol Kim
2020 Sensors  
Our framework provides a logging architecture based on formal specifications, which hierarchically organizes them to find out a priori relationships between them.  ...  However, a test in a situation of certain parameters, so called an oracle test, does not necessarily trigger latent violations of the relationships.  ...  However, in certain situations, oracles do not exist or they are too difficult to apply. This is known as the oracle problem [20] . There is a program in which the test oracle cannot exist.  ... 
doi:10.3390/s20051356 pmid:32121632 fatcat:lbuhrsek5jcazk3n7kggk6hpvy

Machine Learning meets Quantum Foundations: A Brief Survey [article]

Kishor Bharti, Tobias Haug, Vlatko Vedral, Leong-Chuan Kwek
2020 arXiv   pre-print
The goal of machine learning is to facilitate a computer to execute a specific task without explicit instruction by an external party.  ...  Quantum foundations seeks to explain the conceptual and mathematical edifice of quantum theory.  ...  Determining whether an assemblage admits LHS model is a semidefinite program (SDP).  ... 
arXiv:2003.11224v2 fatcat:nstwl5zpurag5lddtfoasrwbja

Beyond Physics? On the Prospects of Finding a Meaningful Oracle

Taner Edis, Maarten Boudry
2014 Foundations of Science  
Examining how a candidate for such an oracle would be evaluated leads to questions that do not admit an easy resolution.  ...  Since we lack any plausible candidate for any such oracle, however, chance-and-necessity physicalism appears very likely to be correct.  ...  Acknowledgments The authors would like to thank Peter Verdée for sharing his expertise on computability and oracles, and Yon Fishman as well as two anonymous reviewers for their insightful comments on  ... 
doi:10.1007/s10699-014-9349-z fatcat:juqvbuwpdnb6hh6scrri67azui

Interactive foundations of computing

Peter Wegner
1998 Theoretical Computer Science  
Computable functions cannot model real-world behavior because functions are too strong an abstraction, sacrificing the ability to model time and other real-world properties to realize formal tractability  ...  The claim that interactive systems have richer behavior than algorithms is surprisingly easy to prove.  ...  distributed artijicial intelligence (agent-oriented programming) > logic-based AI The paradigm shift in AI from logic and search to interactive models is not merely a tactical change but is a strategic  ... 
doi:10.1016/s0304-3975(97)00154-0 fatcat:jfv7xn5nlveqjagut2rkidjdem

Conservative and Adaptive Penalty for Model-Based Safe Reinforcement Learning

Yecheng Jason Ma, Andrew Shen, Osbert Bastani, Jayaraman Dinesh
2022 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
However, imperfect models can result in real-world constraint violations even for actions that are predicted to satisfy all constraints.  ...  Model-based RL algorithms hold promise for reducing unsafe real-world actions: they may synthesize policies that obey all constraints using simulated samples from a learned model.  ...  Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein  ... 
doi:10.1609/aaai.v36i5.20478 fatcat:3gcryjg2qjfhxbmgtlpo52bqsm

The Computational and Pragmatic Approach to the Dynamics of Science

Witold Marciszewski
2018 Filozofia i Nauka  
This essay suggests considering Turing's idea of oracle as a complementary model to explain most adequately, in terms of exceptional inventiveness, the dynamics of mathematics and mathematizable empirical  ...  The computational approach in problem-solving processes consists in considering them in terms of computability: either as being performed according to a model of computation in a narrower sense, e.g.,  ...  In contrast, the UTM model is not extensionally equivalent to the analogcontinuous model of computation (described by means of real recursive functions).  ... 
doi:10.37240/fin.2020.8.1.2 fatcat:rgtgc5t3ejegnl2acnpykycypm

Guided Learning of Control Graphs for Physics-Based Characters

Libin Liu, Michiel Van De Panne, Kangkang Yin
2016 ACM Transactions on Graphics  
The output is a control graph that supports real-time physics-based simulation of multiple characters, each capable of a diverse range of robust movement skills, such as walking, running, sharp turns,  ...  This leverages the results of open-loop sampling-based reconstruction in order to produce state-action pairs which are then transformed into a linear feedback policy for each control fragment using linear  ...  ACKNOWLEDGMENTS This project is partially supported by NSERC Discovery Grants Program RGPIN-2015-04843 and Singapore Ministry of Education Academic Research Fund, Tier 2 (MOE2011-T2-2-152).  ... 
doi:10.1145/2893476 fatcat:vshdaunjbrdxvgraskhdaum3uu

A Data-Driven Approach to Dynamically Adjust Resource Allocation for Compute Clusters [article]

Francesco Pace, Dimitrios Milios, Damiano Carra, Daniele Venzano and Pietro Michiardi
2018 arXiv   pre-print
Nowadays, data-centers are largely under-utilized because resource allocation is based on reservation mechanisms which ignore actual resource utilization.  ...  Our approach monitors resource utilization and employs a data-driven approach to resource demand forecasting, featuring quantification of uncertainty in the predictions.  ...  Oracle-based resource shaping.  ... 
arXiv:1807.00368v1 fatcat:7byxnyhnkjatvnx4hio5qvg2e4

What Would Jiminy Cricket Do? Towards Agents That Behave Morally [article]

Dan Hendrycks, Mantas Mazeika, Andy Zou, Sahil Patel, Christine Zhu, Jesus Navarro, Dawn Song, Bo Li, Jacob Steinhardt
2022 arXiv   pre-print
In extensive experiments, we find that the artificial conscience approach can steer agents towards moral behavior without sacrificing performance.  ...  To facilitate the development of agents that avoid causing wanton harm, we introduce Jiminy Cricket, an environment suite of 25 text-based adventure games with thousands of diverse, morally salient scenarios  ...  This shows that steering behavior with language models possessing ethical understanding is a promising way to tackle the problem of safe exploration.  ... 
arXiv:2110.13136v2 fatcat:aues5yneqvd2lckpx7afdftg44

A Tour of Reinforcement Learning: The View from Continuous Control [article]

Benjamin Recht
2018 arXiv   pre-print
behavior.  ...  In particular, theory and experiment demonstrate the role and importance of models and the cost of generality in reinforcement learning algorithms.  ...  First, this work was generously supported in part by two forward-looking programs at DOD: the Mathematical Data Science program at ONR and the Foundations and Limits of Learning program at DARPA.  ... 
arXiv:1806.09460v2 fatcat:4ago566hwzcudj3mcbb3j5wyr4

Conservative and Adaptive Penalty for Model-Based Safe Reinforcement Learning [article]

Yecheng Jason Ma, Andrew Shen, Osbert Bastani, Dinesh Jayaraman
2021 arXiv   pre-print
However, imperfect models can result in real-world constraint violations even for actions that are predicted to satisfy all constraints.  ...  Model-based RL algorithms hold promise for reducing unsafe real-world actions: they may synthesize policies that obey all constraints using simulated samples from a learned model.  ...  Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein  ... 
arXiv:2112.07701v1 fatcat:fpaqm7dd5jfujcwzozgn4yotly

A Hierarchical MPC Approach to Car-Following via Linearly Constrained Quadratic Programming [article]

Fangyu Wu, Alexandre Bayen
2022 arXiv   pre-print
In this article, we design a novel non-local tri-layer MPC controller that is capable of smoothing a wide range of oscillatory traffic and is amenable to real-time applications.  ...  At the core of the controller design are 1) an accessible prediction method based on ETA estimation and 2) a robust, light-weight optimization procedure, designed specifically for handling various headway  ...  Nevertheless, we know from common sense that such behavior is not acceptable in most of the real-world applications.  ... 
arXiv:2205.10781v2 fatcat:jlqollty5jb5bpvmowf5vzmydi

Game Theory for Unmanned Vehicle Path Planning in the Marine Domain: State of the Art and New Possibilities

Marco Cococcioni, Lorenzo Fiaschi, Pierre F. J. Lermusiaux
2021 Journal of Marine Science and Engineering  
Thanks to the advent of new technologies and higher real-time computational capabilities, the use of unmanned vehicles in the marine domain has received a significant boost in the last decade.  ...  To achieve autonomy in such highly dynamic uncertain conditions, many types of autonomous path planning problems need to be solved.  ...  response model [12, 13] of the adversary's behavior (which is known to better model human-based decision making processes), and it leverages a compact representation of the defender's strategy space  ... 
doi:10.3390/jmse9111175 fatcat:3v6oyeo4yfgcvatahm5rki456u
« Previous Showing results 1 — 15 out of 780 results