4,983 Hits in 3.9 sec

Virtual Screening on FPGA

Tom Vander Aa, Tom Ashby, Roel Wuyts
2019 Belgium-Netherlands Conference on Artificial Intelligence  
Figure 1 shows a simplified view on the prediction flow for activity prediction. A molecule represented by its chemical fingerprint is fed in on the left.  ...  From the fingerprint an internal latent representation is computed and this latent representation is used to make predictions on one or more protein targets.  ... 
dblp:conf/bnaic/AaAW19 fatcat:s2pxvarnnbf5pmhlpfemhj6ig4

Privacy Norms in Online Social Networks

Onuralp Ulusoy, Pinar Yolum
2019 Belgium-Netherlands Conference on Artificial Intelligence  
One of our future goals is to study dynamism in norms.  ...  previous privacy decisions that are taken by the users on a given content type.  ... 
dblp:conf/bnaic/UlusoyY19 fatcat:wpaabfdv6bgdbll3to4epsbbse

An Interactive Consultant

Pierre Carbonnelle, Bram Aerts, Marjolein Deryck, Joost Vennekens, Marc Denecker
2019 Belgium-Netherlands Conference on Artificial Intelligence  
"Augmented Intelligence" describes systems where human and machine work together and learn how to solve a problem by taking advantage of their respective strengths.  ...  In future work, funded by the Flemish Impulsprogram on AI, we plan to investigate how to guide the conversation with the user to solve a specific (sub-)goal, how to compute a recommendation based on uncertain  ... 
dblp:conf/bnaic/CarbonnelleADVD19 fatcat:er5ssjae2zb4zpwwepngvihdwu

SubSect - An Interactive Itemset Visualization

Joey De Pauw, Sandy Moens, Bart Goethals
2019 Belgium-Netherlands Conference on Artificial Intelligence  
Our visualization is based on the double decker plot from Hofmann et al.  ...  A live version with examples can be found on  ... 
dblp:conf/bnaic/PauwMG19 fatcat:nxim2rzvuzh4hpwbsdurkt5kue

On Domination and Control in Strategic Ability

Damian Kurpiewski, Michal Knapik, Wojciech Jamroga
2019 Belgium-Netherlands Conference on Artificial Intelligence  
We prove that the notion of dominance based on the comparison of input/output characteristics is sound, i.e., a dominating strategy can achieve at least what the dominated one can.  ...  We also note that our new approach can handle models that do not submit to the fixpoint approximation scheme (i.e., Castles), as well as ones on which the output of SMC is faulty (i.e., Bridge Endplay)  ... 
dblp:conf/bnaic/KurpiewskiKJ19 fatcat:cbl3y6dfjnabxpts6i3fuufsci

Iterative Model-Based Transfer in Deep Reinforcement Learning

Jelmer Neeven
2019 Belgium-Netherlands Conference on Artificial Intelligence  
In recent years, advances in the field of Deep Reinforcement Learning (DRL) have enabled artificial agents to obtain superhuman performance on various tasks, given enough interactions with the environment  ...  While the state of the art in DRL keeps improving rapidly, most algorithms result in agents that generalize badly, performing well only on the single task they were trained on [9] .  ... 
dblp:conf/bnaic/Neeven19 fatcat:qeg6el7vqjbixecxmbjgcqtluu

Deep Learning Applied to Sign Language

Jérôme Fink, Anthony Cleve, Benoît Frénay
2019 Belgium-Netherlands Conference on Artificial Intelligence  
Usually, one network focuses on the temporal while the other extracts 2D information from one particular frame.  ...  The angle of the camera hides one finger in the example for the number 5. Preliminary experiments have only considered one frame for each video.  ... 
dblp:conf/bnaic/FinkCF19 fatcat:iwecohaeqnegrpx34tudpl7f64

Towards Deterministic Diverse Subset Sampling

Joachim Schreurs, Michaël Fanuel, Johan A. K. Suykens
2019 Belgium-Netherlands Conference on Artificial Intelligence  
The selected landmark points are visualized on the left, the projection on the 2 first principal components of the KPCA on the right.  ...  We demonstrate the use of the proposed method on a two image summarization tasks. An example is given on Figure 1 . The dataset has 9 classes consisting of each 11 images.  ... 
dblp:conf/bnaic/SchreursFS19 fatcat:g623eiklcrhgfm7m2bzgt6c45q

Interpreting and Explaining Deep Models Visually

José Oramas M., Kaili Wang, Tinne Tuytelaars
2019 Belgium-Netherlands Conference on Artificial Intelligence  
Methods based on deep neural networks (DNNs) have achieved impressive results for several computer vision tasks, such as image classification, object detection, and image generation, etc.  ...  Combined with the general tendency in the community of developing methods with a focus on high quantitative performance, this has motivated the wide adoption of DNN-based methods, despite the initial skepticism  ...  As shown on the image below, we produce average visualizations of these features to enable visual interpretation of the model.  ... 
dblp:conf/bnaic/MWT19 fatcat:bsutgtcxhffddby7auapba3m64

Nonlinear Causality Inference in Microarray Time Series

Fateme Nateghi Haredasht, Mohammad Hassan Moradi
2019 Belgium-Netherlands Conference on Artificial Intelligence  
One of the most popular causality inference methods is Granger causality that is a linear regression-based model for determining whether a single time series is useful in forecasting another; however,  ...  One of the advantages of using Granger causality is to infer causal networks like gene regulatory networks (GRNs) from microarray time series data [1, 5, 3] .  ... 
dblp:conf/bnaic/HaredashtM19 fatcat:wxhvx44z25c25b7ziz77xe344m

SOGrounder: Modelling and Solving Second-Order Logic

Matthias van der Hallen, Gerda Janssens
2019 Belgium-Netherlands Conference on Artificial Intelligence  
Introduction A knowledge representation (KR) system allows declarative modelling of realworld knowledge, supporting one or more inferences to be performed on the model, e.g. satisfiability checking, model  ...  In previous work [1] , this system was tested with different QBF solvers on the strategic companies problem, a well-known problem in ASP and QBF communities.  ... 
dblp:conf/bnaic/HallenJ19 fatcat:p5m3foy7gndbze6go5ukttf2i4

Explaining Lexical Processing Times with Cognitively Plausible Computational Models

Wietse de Vries
2019 Belgium-Netherlands Conference on Artificial Intelligence  
Results The trained models are evaluated on the POS tags of documents in the PROVO corpus.  ...  To explain variance in human prediction errors with model results, ordinary least sequares (OLS) models are fit on the results.  ... 
dblp:conf/bnaic/Vries19 fatcat:nrfymrv2z5aevjjillrnyt4q7e

Integrating Clinically-Relevant Features into Skin Lesion Classification

Emmeke Veltmeijer, Sezer Karaoglu, Theo Gevers
2019 Belgium-Netherlands Conference on Artificial Intelligence  
Introduction Skin cancer is one of the most common types of cancer, melanoma being the most deadly form of skin cancer [3] .  ...  This information is then provided to the network as auxiliary output to train on, next to the main ground truth classification (classifying malignant melanoma and two benign skin lesion types).  ... 
dblp:conf/bnaic/VeltmeijerKG19 fatcat:4v2avxaixzdqbizphxzoofxexa

A Virtual Maze Game to Explain Reinforcement Learning

Youri Coppens, Eugenio Bargiacchi, Ann Nowé
2019 Belgium-Netherlands Conference on Artificial Intelligence  
This application visualises the learning process of Watkins' Q(λ), one of the fundamental algorithms in the field. A video can be found at  ...  The user can freely select actions and decide where to explore depending on the available information.  ...  As the player explores, Q(λ) updates the Q-values for each state-action pair and displays them on the ground.  ... 
dblp:conf/bnaic/CoppensBN19 fatcat:ogtr6ljrbreyjdlldbubawoidy

News Topic Recommendation Using an Extended Bayesian Personalized Ranking

Alireza Gharahighehi, Celine Vens
2019 Belgium-Netherlands Conference on Artificial Intelligence  
The extended version of BPR performs better compare to the original version based on two evaluation measures.  ...  Bayesian Personalized Ranking (BPR) is a recommendation approach which learns to rank candidate items based on user's implicit feedback.  ...  For instance, when a user visits a web-page or clicks on a link or adds an item to her basket the feedback is implicit.  ... 
dblp:conf/bnaic/GharahighehiV19 fatcat:5bwscppia5fzzp74enrzyo7jia
« Previous Showing results 1 — 15 out of 4,983 results