Filters








46,637 Hits in 3.1 sec

Explaining Deep Learning Models using Causal Inference [article]

Tanmayee Narendra, Anush Sankaran, Deepak Vijaykeerthy, Senthil Mani
2018 arXiv   pre-print
In this work, we use ideas from causal inference to describe a general framework to reason over CNN models.  ...  Although deep learning models have been successfully applied to a variety of tasks, due to the millions of parameters, they are becoming increasingly opaque and complex.  ...  (Kusner et al. 2017) Ideas from causal inference have also been used in providing human understandable explanations in deep learning models.  ... 
arXiv:1811.04376v1 fatcat:q3t2jhgq4fhonoinwjaq2frv2y

Knowledge-powered Explainable Artificial Intelligence (XAI) for Network Automation Towards 6G [article]

Yulei Wu and Guozhi Lin and Jingguo Ge
2022 arXiv   pre-print
Deep learning models developed to enable network automation for given operation practices have the limitations of 1) lack of explainability and 2) inapplicable across different networks and/or network  ...  To tackle the above issues, in this article we propose a new knowledge-powered framework that provides a human-understandable explainable artificial intelligence (XAI) agent for network automation.  ...  Deep learning models to solve traditional methods Deep learning model explainers Deep learning / transfer learning models Deep learning model explainers Our work A knowledge-powered human-understandable  ... 
arXiv:2205.05406v1 fatcat:rfldsm3lfffr7mors3dcbjxxsa

Leveraging Causal Inference for Explainable Automatic Program Repair [article]

Jianzong Wang, Shijing Si, Zhitao Zhu, Xiaoyang Qu, Zhenhou Hong, Jing Xiao
2022 arXiv   pre-print
Deep learning models have made significant progress in automatic program repair. However, the black-box nature of these methods has restricted their practical applications.  ...  To address this challenge, this paper presents an interpretable approach for program repair based on sequence-to-sequence models with causal inference and our method is called CPR, short for causal program  ...  For all datasets, causal inference is used to improve the effect of debugging code. We test in four languages, and both models have been improved using causal inference.  ... 
arXiv:2205.13342v1 fatcat:o6jij3ack5abtbfdoqahruv7fq

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
Therefore, interpreting machine learning model is currently a mainstream topic in the research community.  ...  However, the traditional interpretable machine learning focuses on the association instead of the causality.  ...  OPEN QUESTIONS AND DISCUSSIONS The need of explaining and interpreting models becomes highly critical along with the growing popularity of deep learning and automated machine learning.  ... 
arXiv:2006.16789v2 fatcat:ole3dvpnjnfkflldd6to4nrrwq

When Causal Intervention Meets Adversarial Examples and Image Masking for Deep Neural Networks [article]

Chao-Han Huck Yang, Yi-Chieh Liu, Pin-Yu Chen, Xiaoli Ma, Yi-Chang James Tsai
2019 arXiv   pre-print
Discovering and exploiting the causality in deep neural networks (DNNs) are crucial challenges for understanding and reasoning causal effects (CE) on an explainable visual model.  ...  "Intervention" has been widely used for recognizing a causal relation ontologically. In this paper, we propose a causal inference framework for visual reasoning via do-calculus.  ...  Ideas from causal inference have also been used in providing comprehensive explanations of deep learning models.  ... 
arXiv:1902.03380v3 fatcat:duh3oiyxazbmrjwmlkfoi2xll4

Beauty Learning and Counterfactual Inference [article]

Tao Li
2019 arXiv   pre-print
We introduce the beauty learning problem as an example, which has been discussed metaphysically for centuries and been proved exists, is quantifiable, and can be learned by deep models in our recent paper  ...  We expect the proposed framework for a broader application in causal inference.  ...  Structural Causal Model Structural Causal Model (SCM) [4] lies in the key of structural causal inference.  ... 
arXiv:1904.12629v1 fatcat:lz3oavltxjdefal5tep7bix6nm

Machine Learning and Deep Learning – A review for Ecologists [article]

Maximilian Pichler, Florian Hartig
2022 arXiv   pre-print
The popularity of Machine learning (ML), Deep learning (DL), and Artificial intelligence (AI) has sharply risen in recent years.  ...  Recently, however, they have been increasingly used for classical analytical tasks traditionally covered by statistical models.  ...  On the other hand, their use for causal inference is still disputed.  ... 
arXiv:2204.05023v1 fatcat:ft3nggdzc5fhtg4hsu2rgw45sm

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

Raha Moraffah, Mansooreh Karami, Ruocheng Guo, Adrienne Raglin, Huan Liu
2020 arXiv   pre-print
Machine learning models have had discernible achievements in a myriad of applications. However, most of these models are black-boxes, and it is obscure how the decisions are made by them.  ...  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.  ...  Causal Inference and Model-based Interpretation Recently, causality has gained increasing attention in explaining machine learning models [12; 38] .  ... 
arXiv:2003.03934v3 fatcat:awzv47nmv5aqtl4j5asmp5v7zq

Interpretable and Explainable Machine Learning for Materials Science and Chemistry [article]

Felipe Oviedo, Juan Lavista Ferres, Tonio Buonassisi, Keith Butler
2021 arXiv   pre-print
In particular, we emphasize the risks of inferring causation or reaching generalization by purely interpreting machine learning models and the need of uncertainty estimates for model explanations.  ...  The predictions and inner workings of models should provide a certain degree of explainability by human experts, permitting the identification of potential model issues or limitations, building trust on  ...  ACKNOWLEDGEMENTS We thank Professor Volker Deringer and Dr Noor Titan Putri Hartono for useful discussion. We thank Pedro Costa for his contributions to figure design.  ... 
arXiv:2111.01037v2 fatcat:hirrciqrlbfcfdmo4bi4sv4lxm

Deep Structural Causal Models for Tractable Counterfactual Inference [article]

Nick Pawlowski, Daniel C. Castro, Ben Glocker
2020 arXiv   pre-print
We formulate a general framework for building structural causal models (SCMs) with deep learning components.  ...  deep causal learning methods.  ...  We call an SCM that uses deep-learning components to model the structural assignments a deep structural causal model (DSCM).  ... 
arXiv:2006.06485v2 fatcat:tysaxdgjd5avvm5l7ovdkimhce

Building machines that learn and think like people

Brenden M. Lake, Tomer D. Ullman, Joshua B. Tenenbaum, Samuel J. Gershman
2016 Behavioral and Brain Sciences  
Specifically, we argue that these machines should (1) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (2) ground learning  ...  Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats that of humans  ...  Tom Schaul was very helpful in answering questions regarding the DQN learning curves and Frostbite scoring.  ... 
doi:10.1017/s0140525x16001837 pmid:27881212 fatcat:3fjriprksbhaxpqdcydrhmcjqm

Building Machines That Learn and Think Like People [article]

Brenden M. Lake, Tomer D. Ullman, Joshua B. Tenenbaum, Samuel J. Gershman
2016 arXiv   pre-print
Specifically, we argue that these machines should (a) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (b) ground learning  ...  Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats humans in some  ...  Tom Schaul was very helpful in answering questions regarding the DQN learning curves and Frostbite scoring.  ... 
arXiv:1604.00289v3 fatcat:ph2rrwk2znb4dpb5nvcg54x2xi

A multi-component framework for the analysis and design of explainable artificial intelligence [article]

S. Atakishiyev, H. Babiker, N. Farruque, R. Goebel1, M-Y. Kima, M.H. Motallebi, J. Rabelo, T. Syed, O. R. Zaïane
2020 arXiv   pre-print
First, the enormous application success of modern machine learning methods, especially deep and reinforcement learning, which have created high expectations for industrial, commercial and social value.  ...  The rapid growth of research in explainable artificial intelligence (XAI) follows on two substantial developments.  ...  As examples of the explainable model, there can be a causal model, an explainable deep adaptive program, an explainable reinforcement learning model, etc.  ... 
arXiv:2005.01908v1 fatcat:af7fhjzyhveellylg3zqdtobpm

Causal Reasoning Meets Visual Representation Learning: A Prospective Study [article]

Yang Liu, Yushen Wei, Hong Yan, Guanbin Li, Liang Lin
2022 arXiv   pre-print
Inspired by the strong inference ability of human-level agents, recent years have therefore witnessed great effort in developing causal reasoning paradigms to realize robust representation and model learning  ...  In this paper, we conduct a comprehensive review of existing causal reasoning methods for visual representation learning, covering fundamental theories, models, and datasets.  ...  Causal Visual Robustness The ubiquitous spurious correlation learned by deep learning models reduces the model robustness, which is a potential vulnerability of the conventional deep learning paradigm.  ... 
arXiv:2204.12037v4 fatcat:fgrz2vh42bdozcb5lfegfv5x34

Explainable and Interpretable Deep Learning Models

Md Shamsuzzaman
2020 Global Journal of Engineering Sciences  
Machine learning (ML) and deep learning (DL) are branches of artificial intelligence, where the model with the associated parameters are developed using data.  ...  Opinion Deep learning models are becoming ubiquitous recently. However, they still suffer from many limitations. In this mini review, we will focus on one of these aspects of deep learning models.  ...  model to learn an explainable, causal, probabilistic programming model; pattern theory; adaptive programs; cognitive models; reinforcement and attention-based learning; question-answer system; incorporating  ... 
doi:10.33552/gjes.2020.05.000621 fatcat:uwqj6bfalbc7vh4h7hjzcru2mu
« Previous Showing results 1 — 15 out of 46,637 results