Filters








6,525 Hits in 4.1 sec

Embedding Normative Reasoning into Neural Symbolic Systems

Guido Boella, Silvano Colombo Tosatto, Artur S. d'Avila Garcez, Valerio Genovese, Leon van der Torre
2011 International Workshop on Neural-Symbolic Learning and Reasoning  
Considering this problem, we propose a neuralsymbolic approach to provide agents the instruments to reason about and learn norms in a dynamic environment.  ...  The resulting system called Normative-CILP(N-CILP) shows how neural networks can cope with some of the underpinnings of normative reasoning: permissions, dilemmas, exceptions and contrary to duty problems  ...  Figure 1 : 1 Figure 1: Neural-Symbolic Normative Agent. Figure 2 : 2 Figure 2: Neural-Symbolic simulator.  ... 
dblp:conf/nesy/BoellaTGGT11 fatcat:qgctp3okwfhzdpttqn27yqm6qm

Top-Down and Bottom-Up Interactions between Low-Level Reactive Control and Symbolic Rule Learning in Embodied Agents

Clément Moulin-Frier, Xerxes D. Arsiwalla, Jordi-Ysard Puigbo, Martí Sánchez-Fibla, Armin Duff, Paul F. M. J. Verschure
2016 Neural Information Processing Systems  
symbolic rule learning in embodied agents.  ...  The interaction of these modules in a closed-loop fashion suggests how symbolic representations might have been shaped from low-level behaviors and recruited for behavior optimization.  ...  The research direction we adopt is based on the hypothesis that social norms are needed for the evolution of large multi-agent groups and that the formation of those social norms requires each individual  ... 
dblp:conf/nips/Moulin-FrierAPS16 fatcat:zepcxsghffhcvgiigrowskpfee

Neural Abstract Reasoner [article]

Victor Kolev, Bogdan Georgiev, Svetlin Penkov
2020 arXiv   pre-print
We introduce the Neural Abstract Reasoner (NAR), a memory augmented architecture capable of learning and using abstract rules.  ...  reasoning and logic inference are difficult problems for neural networks, yet essential to their applicability in highly structured domains.  ...  Related Work Neuro-symbolic architectures Hybrid neuro-symbolic approaches enable agents to solve structured tasks from raw data, while learning faster and being more robust to noise [9, 28, 19, 16] .  ... 
arXiv:2011.09860v1 fatcat:6ayeada5hvg27d37d3jvzoj6le

Review of Trust and Machine Ethics Research: Towards A Bio-Inspired Computational Model of Ethical Trust (CMET)

Hock Chuan Lim, Rob Stocker, Henry Larkin
2008 Proceedings of the Third International Conference on Bio-Inspired Models of Network Information and Computing Systems (Bionetics 2008)  
Recent advances in the fields of robotics, cyborg development, moral psychology, trust, multi agent-based systems and socionics have raised the need for a better understanding of ethics, moral reasoning  ...  of collective social moral norms.  ...  Key features of the CMET are ethical trust reasoning and bio-inspired neural agent-based processes in the evolution of moral norms.  ... 
doi:10.4108/icst.bionetics2008.4728 dblp:conf/bionetics/LimSL08 fatcat:cqpncqlagvcg3kv2ew2tfd5ylm

Logic-Based Technologies for Intelligent Systems: State of the Art and Perspectives

Roberta Calegari, Giovanni Ciatto, Enrico Denti, Andrea Omicini
2020 Information  
Together with the disruptive development of modern sub-symbolic approaches to artificial intelligence (AI), symbolic approaches to classical AI are re-gaining momentum, as more and more researchers exploit  ...  Future perspectives for exploitation of logic-based technologies are discussed as well, in order to identify those research fields that deserve more attention, considering the areas that already exploit  ...  It consists of an interdisciplinary effort combining methods and results from several sources, from deontic logic, norms and agent-based simulation to game theory and norms, normative agents, norms and  ... 
doi:10.3390/info11030167 fatcat:e3wed54dyzabldrnml7khx37te

Dynamic Cognition Applied to Value Learning in Artificial Intelligence [article]

Nythamar de Oliveira, Nicholas Kluge Corrêa
2021 arXiv   pre-print
Nevertheless, if such an advance isn't done with prudence, it can result in negative outcomes for humanity.  ...  For this reason, several researchers in the area are trying to develop a robust, beneficial, and safe concept of artificial intelligence.  ...  The limits imposed by symbolic architecture are another reason for criticism of the symbolic method.  ... 
arXiv:2005.05538v5 fatcat:ob5rvlejxna57lra5fsc5c6mzy

Symbolic state transducers and recurrent neural preference machines for text mining

Garen Arevian, Stefan Wermter, Christo Panchev
2003 International Journal of Approximate Reasoning  
These encoding symbolic transducers and learning neural preference machines can be seen as independent agents, each one tackling the same task in a different manner.  ...  An experimental analysis of the performance of these symbolic transducer and neural preference machines is presented.  ...  This work on neural agents [40] [41] [42] [43] has been demonstrated for the task of textual classification.  ... 
doi:10.1016/s0888-613x(02)00085-3 fatcat:evzt5r4g7fg2henpjrnetkg6wi

Catformer: Designing Stable Transformers via Sensitivity Analysis

Jared Quincy Davis, Albert Gu, Krzysztof Choromanski, Tri Dao, Christopher Ré, Chelsea Finn, Percy Liang
2021 International Conference on Machine Learning  
Sensitivity characterizes how the variance of activation and gradient norms change in expectation when parameters are randomly perturbed.  ...  We analyze the sensitivity of previous Transformer architectures and design a new architecture, the Catformer, which replaces residual connections or RNN-based gating mechanisms with concatenation.  ...  of pre-norm architectures.  ... 
dblp:conf/icml/DavisGCDRFL21 fatcat:ldli67dr2fgabjaz6wz6jtneui

Human-Like Neural-Symbolic Computing (Dagstuhl Seminar 17192)

Tarek R. Besold, Artur D'Avila Garcez, Luis C. Lamb, Marc Herbstritt
2017 Dagstuhl Reports  
previous Dagstuhl Seminar 14381: Neural-Symbolic Learning and Reasoning.  ...  It was built upon previous seminars and workshops on the integration of computational learning and symbolic reasoning, such as the Neural-Symbolic Learning and Reasoning (NeSy) workshop series, and the  ...  Recently, many have rejected the interest of the syllogism task and argued for probability as the appropriate normative framework for human reasoning.  ... 
doi:10.4230/dagrep.7.5.56 dblp:journals/dagstuhl-reports/BesoldGL17 fatcat:xf66ju6bhrdynclruq4zidd2xi

Reasoning in Non-Probabilistic Uncertainty: Logic Programming and Neural-Symbolic Computing as Examples [article]

Tarek R. Besold, Artur d'Avila Garcez, Keith Stenning, Leendert van der Torre, Michiel van Lambalgen
2017 arXiv   pre-print
of the intended model, and a neural-symbolic implementation of Input/Output logic for dealing with uncertainty in dynamic normative contexts.  ...  For the latter claim, two paradigmatic examples are presented: Logic Programming with Kleene semantics for modelling reasoning from information in a discourse, to an interpretation of the state of affairs  ...  Acknowledgements We want to thank the following people for their indispensable contributions to different parts of the work reported in this article: Guido Boella, Silvano Colombo Tosatto, Valerio Genovese  ... 
arXiv:1701.05226v2 fatcat:eg67ppsao5cmvnxv5s5vquchma

Reasoning in Non-probabilistic Uncertainty: Logic Programming and Neural-Symbolic Computing as Examples

Tarek R. Besold, Artur d'Avila Garcez, Keith Stenning, Leendert van der Torre, Michiel van Lambalgen
2017 Minds and Machines  
of the intended model, and a neural-symbolic implementation of input/output logic for dealing with uncertainty in dynamic normative contexts.  ...  For the latter claim, two paradigmatic examples are presented: logic programming with Kleene semantics for modelling reasoning from information in a discourse, to an interpretation of the state of affairs  ...  Acknowledgements We want to thank the following people for their indispensable contributions to different parts of the work reported in this article: Guido Boella, Silvano Colombo Tosatto, Valerio Genovese  ... 
doi:10.1007/s11023-017-9428-3 fatcat:unbwsv3civd45hivgh5xpuzlzu

Dynamic Models Applied to Value Learning in Artificial Intelligence

Nythamar De Oliveira, Nicholas Kluge Corrêa
2020 Figshare  
want such systems to do, especially when we look for the possibility of intelligent agents acting in several domains over the long term.  ...  For this reason, several researchers in the area are trying to develop a robust, beneficial, and safe concept of AI for the preservation of humanity and the environment.  ...  The limits imposed by symbolic architecture are another source of criticism of the computational method.  ... 
doi:10.6084/m9.figshare.12311816.v2 fatcat:6mk3ufd5rvcfxc6p3pphhdscaa

Planning with iFALCON: Towards A Neural-Network-Based BDI Agent Architecture

Budhitama Subagdja, Ah-Hwee Tan
2008 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology  
This paper presents iFALCON, a model of BDI (beliefdesire-intention) agents that is fully realized as a selforganizing neural network architecture.  ...  Based on multichannel network model called fusion ART, iFALCON is developed to bridge the gap between a self-organizing neural network that autonomously adapts its knowledge and the BDI agent model that  ...  ≡ min(p i , q i ), and the norm |.| is defined by |p| ≡ i p i for vectors p and q.  ... 
doi:10.1109/wiiat.2008.29 dblp:conf/iat/SubagdjaT08 fatcat:hkabflyo25c45kfzesidkzzcoi

Computational Foundations of Natural Intelligence [article]

Marcel van Gerven
2017 bioRxiv   pre-print
Subsequently, I concentrate on the use of artificial neural networks as a framework for modeling cognitive processes.  ...  This paper reviews some of the computational principles relevant for understanding natural intelligence and, ultimately, achieving strong AI.  ...  Acknowledgements This work is supported by a VIDI grant (639.072.513) from the Netherlands Organization for Scientific Research.  ... 
doi:10.1101/166785 fatcat:n347rhnrards7oz352nkuscjb4

T-Norms Driven Loss Functions for Machine Learning [article]

Giuseppe Marra, Francesco Giannini, Michelangelo Diligenti, Marco Maggini, Marco Gori
2020 arXiv   pre-print
This paper shows that the loss function expressing these neural-symbolic learning tasks can be unambiguously determined given the selection of a t-norm generator.  ...  Neural-symbolic approaches have recently gained popularity to inject prior knowledge into a learner without requiring it to induce this knowledge from data.  ...  Related Works Neural-symbolic approaches (Besold et al., 2017; Garcez et al., 2012) aim at combining symbolic reasoning into (deep) neural networks.  ... 
arXiv:1907.11468v4 fatcat:qgsavkwazfgc7ebkpfobuezkqa
« Previous Showing results 1 — 15 out of 6,525 results