The analogy between Gaussian processes (GPs) and deep artificial neural networks (ANNs) has received a lot of interest, and has shown promise to unbox the blackbox of deep ANNs. Existing theoretical works put strict assumptions on the ANN (e.g. requiring all intermediate layers to be wide, or using specific activation functions). Accommodating those theoretical assumptions is hard in recent deep architectures, and those theoretical conditions need refinement as new deep architectures emerge. In this paper we derive an evidence lower-bound that encourages the GP's posterior to match the ANN's output without any requirement on the ANN. Using our method we find out that on 5 datasets, only a subset of those theoretical assumptions are sufficient. Indeed, in our experiments we used a normal ResNet-18 or feed-forward backbone with a single wide layer in the end. One limitation of training GPs is the lack of scalability with respect to the number of inducing points. We use novel computational techniques that allow us to train GPs with hundreds of thousands of inducing points and with GPU acceleration. As shown in our experiments, doing so has been essential to get a close match between the GPs and the ANNs on 5 datasets. We implement our method as a publicly available tool called GPEX: https://github.com/amirakbarnejad/gpex. On 5 datasets (4 image datasets, and 1 biological dataset) and ANNs with 2 types of functionality (classifier or attention-mechanism) we were able to find GPs whose outputs closely match those of the corresponding ANNs. After matching the GPs to the ANNs, we used the GPs' kernel functions to explain the ANNs' decisions. We provide more than 200 explanations (around 30 explanations in the paper and the rest in the supplementary) which are highly interpretable by humans and show the ability of the obtained GPs to unbox the ANNs' decisions.
Hypothesis Testing and Machine Learning: Interpreting Variable Effects in Deep Artificial Neural Networks using Cohen's f2
University of South Carolina(南卡罗来纳大学);University of Kassel(卡塞尔大学);University of Wales(威尔士大学);Technical University Munich(慕尼黑工业大学);University of Newcastle upon Tyne(纽卡斯尔大学);Università per Stranieri di Perugia(佩鲁贾外国人大学)
Interpreting artificial neural networks to detect genome-wide association signals for complex traits
University of Tartu(塔尔图大学);Institute of Genomics(基因组学研究所);Estonian Genome Centre(爱沙尼亚基因组中心);CNRS(法国国家科学研究中心);INRIA(法国国家信息与自动化研究所);LISN(巴黎萨克雷大学国家科学研究实验室);Paris-Saclay University(巴黎萨克雷大学);Estonian Biobank(爱沙尼亚生物银行)
Fraunhofer Heinrich Hertz Institute(弗劳恩霍夫海因里希·赫兹研究所);Technische Universität Berlin(柏林工业大学);Korea University(高丽大学);Max Planck Institute for Informatics(马克斯·普朗克信息学研究所)
Discovering Chunks in Neural Embeddings for Interpretability
Explainable Machine Learning Lab, Helmholtz Munich(慕尼黑亥姆霍兹大学可解释机器学习实验室);Technical University of Munich(慕尼黑工业大学);Institute for Human-Centered AI, Helmholtz Munich(慕尼黑亥姆霍兹以人为本人工智能研究所);Max Planck Institute for Biological Cybernetics(马克斯·普朗克生物控制论研究所);Munich Center for Machine Learning(慕尼黑机器学习中心)
Interpreting Deep Learning Models for Epileptic Seizure Detection on EEG signals
EPFL(洛桑联邦理工学院);University of Applied Sciences Western Switzerland (HEIG-VD/HES-SO)(瑞士西部应用科学大学(日内瓦湖高等学院/瑞士西部应用科学与艺术大学));Centre Hospitalier Universitaire Vaudois (CHUV)(沃州大学中心医院);University of Lausanne(洛桑大学);University of Lyon(里昂大学);Lyon’s Neurosciences Research Center (INSERM U1028/CNRS UMR 5292)(里昂神经科学研究中心(法国国家健康与医学研究院U1028/法国国家科学研究中心UMR 5292))
Artificial Intelligence Algorithms for the Detection of Pathologies Related to Lung Cancer through Image Analysis using Convolutional Neural Networks and Data Augmentation: a systematic mapping of the literature