Filters








485 Hits in 6.9 sec

Causality Learning: A New Perspective for Interpretable Machine Learning [article]

Guandong Xu, Tri Dung Duong, Qian Li, Shaowu Liu, Xianzhi Wang
2021 arXiv   pre-print
The evaluation techniques for assessing method quality, and open problems in causal interpretability are also discussed in this paper.  ...  Therefore, interpreting machine learning model is currently a mainstream topic in the research community.  ...  including graph inference and pairwise inference are provided in Causal Discovery Toolbox.  ... 
arXiv:2006.16789v2 fatcat:ole3dvpnjnfkflldd6to4nrrwq

MARGIN: Uncovering Deep Neural Networks Using Graph Signal Analysis

Rushil Anirudh, Jayaraman J. Thiagarajan, Rahul Sridhar, Peer-Timo Bremer
2021 Frontiers in Big Data  
This paper introduces MARGIN, a simple yet general approach to address a large set of interpretability tasks MARGIN exploits ideas rooted in graph signal analysis to determine influential nodes in a graph  ...  Interpretability has emerged as a crucial aspect of building trust in machine learning systems, aimed at providing insights into the working of complex neural networks that are otherwise opaque to a user  ...  ., 2016) , we use a simple 1-nearest neighbor classifier.  ... 
doi:10.3389/fdata.2021.589417 fatcat:cbequfv24jduxobd3sqiw3jnea

Causal Interpretability for Machine Learning – Problems, Methods and Evaluation [article]

Raha Moraffah, Mansooreh Karami, Ruocheng Guo, Adrienne Raglin, Huan Liu
2020 arXiv   pre-print
In this work, we present a comprehensive survey on causal interpretable models from the aspects of the problems and methods.  ...  or "Was it a specific feature that caused the decision made by the model?". In this work, models that aim to answer causal questions are referred to as causal interpretable models.  ...  [36] provide a comprehensive review of existing causal inference methods and definitions. Definition 1 (Structural Causal Models ).  ... 
arXiv:2003.03934v3 fatcat:awzv47nmv5aqtl4j5asmp5v7zq

Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future [article]

David Ahmedt-Aristizabal, Mohammad Ali Armin, Simon Denman, Clinton Fookes, Lars Petersson
2021 arXiv   pre-print
As such, graph neural networks have attracted significant attention by exploiting implicit information that resides in a biological system, with interactive nodes connected by edges whose weights can be  ...  We provide an overview of these methods in a systematic manner, organized by their domain of application including functional connectivity, anatomical structure and electrical-based analysis.  ...  The simple graph convolution network proposed by Wu et al.  ... 
arXiv:2105.13137v1 fatcat:gm7d2ziagba7bj3g34u4t3k43y

Towards Causal Understanding of Fake News Dissemination [article]

Lu Cheng, Ruocheng Guo, Kai Shu, Huan Liu
2020 arXiv   pre-print
Drawing on theories in causal inference, in this work, we first propose a principled approach to unbiased modelings of fake news dissemination under selection bias.  ...  To mitigate its negative impact, however, we argue that a critical element is to understand why people spread fake news.  ...  A larger value denotes that a sample is further away from its neighboring clusters.  ... 
arXiv:2010.10580v1 fatcat:ewfwojp2zrdpppsapm3vmpiivm

Recommendation system based on heterogeneous feature: A survey

Hui Wang, ZiChun Le, Xuan Gong
2020 IEEE Access  
[22] combined trust information and graph clustering methods in the recommendation system, which expressed the user/item information as a graph.  ...  This method encodes the content's features, time, and trust information into a complex graph on which recommendations are performed using a personalized PageRank.  ... 
doi:10.1109/access.2020.3024154 fatcat:clxk77bcr5hdjd3hnxxi6wzlr4

Neural Graph Embedding Methods for Natural Language Processing [article]

Shikhar Vashishth
2020 arXiv   pre-print
Recently, Graph Convolutional Networks (GCNs) have been proposed to address this shortcoming and have been successfully applied for several problems.  ...  However, a more general and pervasive class of graphs are relational graphs where each edge has a label and direction associated with it.  ...  We propose SynGCN, a Graph Convolution based method for learning word embeddings.  ... 
arXiv:1911.03042v3 fatcat:fruw547yxnev5pmnlij76wovcy

Variant Approach for Identifying Spurious Relations That Deep Learning Models Learn

Tobias Tesch, Stefan Kollet, Jochen Garcke
2021 Frontiers in Water  
A deep learning (DL) model learns a function relating a set of input variables with a set of target variables.  ...  In any case, a description is only useful if one is able to identify if parts of it reflect spurious instead of causal relations (e.g., random associations in the training data instead of associations  ...  Note that in our work the terms causal and spurious do not refer to an underlying causal graph or other concepts from the strict causality framework but should be interpreted with common sense: a pixel  ... 
doi:10.3389/frwa.2021.745563 fatcat:5ftw2ufffnd4jl7bjubxc6uoyq

A Survey of Deep Reinforcement Learning in Recommender Systems: A Systematic Review and Future Directions [article]

Xiaocong Chen, Lina Yao, Julian McAuley, Guanglin Zhou, Xianzhi Wang
2021 arXiv   pre-print
Then, we provide a taxonomy of current DRL-based recommender systems and a summary of existing methods.  ...  In light of the emergence of deep reinforcement learning (DRL) in recommender systems research and several fruitful results in recent years, this survey aims to provide a timely and comprehensive overview  ...  Causal and Counterfactual Inference Causality is a generic relationship between a cause and effect.  ... 
arXiv:2109.03540v2 fatcat:5gwrbfcj3rc7jfkd54eseck5ga

A User Guide to Low-Pass Graph Signal Processing and its Applications [article]

Raksha Ramakrishna, Hoi-To Wai, Anna Scaglione
2020 arXiv   pre-print
What follows is a user guide on a specific class of graph data, where the generating graph filters are low-pass, i.e., the filter attenuates contents in the higher graph frequencies while retaining contents  ...  The notion of graph filters can be used to define generative models for graph data. In fact, the data obtained from many examples of network dynamics may be viewed as the output of a graph filter.  ...  The filter is causal and x s = 0 for s < 0.  ... 
arXiv:2008.01305v1 fatcat:fuexcavtofbk5mxwqbcage7uau

Machine Learning Applications for Therapeutic Tasks with Genomics Data [article]

Kexin Huang, Cao Xiao, Lucas M. Glass, Cathy W. Critchlow, Greg Gibson, Jimeng Sun
2021 arXiv   pre-print
Thanks to the increasing availability of genomics and other biomedical data, many machine learning approaches have been proposed for a wide range of therapeutic discovery and development tasks.  ...  Graph Figure 11: Task illustrations for the theme "inferring causal effects".  ...  causal inference (Davey Smith & Ebrahim 2003).  ... 
arXiv:2105.01171v1 fatcat:d2nbrjt4tvak7momoxxjlmqk2m

Timing Analysis for Inferring the Topology of the Bitcoin Peer-to-Peer Network

Till Neudecker, Philipp Andelfinger, Hannes Hartenstein
2016 2016 Intl IEEE Conferences on Ubiquitous Intelligence & Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People, and Smart World Congress (UIC/ATC/ScalCom/CBDCom/IoP/SmartWorld)  
In this paper we present a timing analysis method that targets flooding P2P networks and show its theoretical and practical feasibility.  ...  A validation in the real-world Bitcoin network proves the possibility of inferring network links of actively participating peers with substantial precision and recall (both ∼ 40 %), potentially enabling  ...  CONCLUSION In this paper we presented a timing analysis method for inferring the topology of flooding P2P networks.  ... 
doi:10.1109/uic-atc-scalcom-cbdcom-iop-smartworld.2016.0070 dblp:conf/uic/NeudeckerAH16 fatcat:mrobknoaa5ab7nymfusn7l4cty

Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future

David Ahmedt-Aristizabal, Mohammad Ali Armin, Simon Denman, Clinton Fookes, Lars Petersson
2021 Sensors  
As such, graph neural networks have attracted significant attention by exploiting implicit information that resides in a biological system, with interacting nodes connected by edges whose weights can be  ...  We provide an overview of these methods in a systematic manner, organized by their domain of application including functional connectivity, anatomical structure, and electrical-based analysis.  ...  [19] suggested a convolution filter on a sphere, termed Direct Neighbor, which is used to develop surface convolution, pooling and transposed convolution in spherical space.  ... 
doi:10.3390/s21144758 fatcat:jytyt4u2pjgvhnhcto3vcvd3a4

No Press Diplomacy: Modeling Multi-Agent Gameplay [article]

Philip Paquette, Yuchen Lu, Steven Bocco, Max O. Smith, Satya Ortiz-Gagne, Jonathan K. Kummerfeld, Satinder Singh, Joelle Pineau, Aaron Courville
2019 arXiv   pre-print
Reliance on trust and coordination makes Diplomacy the first non-cooperative multi-agent benchmark for complex sequential social dilemmas in a rich environment.  ...  Diplomacy is a seven-player non-stochastic, non-cooperative game, where agents acquire resources through a mix of teamwork and betrayal.  ...  Graph Convolution Network with FiLM To take advantage of the adjacency information on the Diplomacy map, we propose to use a graph convolution-based encoder [28] .  ... 
arXiv:1909.02128v2 fatcat:grib4eodlbgq7lqd3sktlpmy3y

VAE-CE: Visual Contrastive Explanation using Disentangled VAEs [article]

Yoeri Poels, Vlado Menkovski
2021 arXiv   pre-print
We build the model using a disentangled VAE, extended with a new supervised method for disentangling individual dimensions.  ...  An analysis on synthetic data and MNIST shows that the approaches to both disentanglement and explanation provide benefits over other methods.  ...  Each step methods. The model described in §3.1 forms the baseline. should indicate a single concept being changed.  ... 
arXiv:2108.09159v1 fatcat:onejltviqzavfm6phethxxux3q
« Previous Showing results 1 — 15 out of 485 results