Toward Anthropomorphic Machine Learning

Jiang

et al. 2021

WIREs Data Min & Knowl

Self Cite

376

115

This paper provides a brief analytical review of the current state-of-the-art in relation to the explainability of artificial intelligence in the context of recent advances in machine learning and deep learning. The paper starts with a brief historical introduction and a taxonomy, and formulates the main challenges in terms of explainability building on the recently formulated National Institute of Standards four principles of explainability. Recently published methods related to the topic are then critically reviewed and analyzed. Finally, future directions for research are suggested.

Section: Brief Historical Perspectivementioning

confidence: 99%

Explainable artificial intelligence: an analytical review

Jiang

et al. 2021

WIREs Data Min & Knowl

Self Cite

376

115

“…The reason it is Cauchy is not arbitrary [4]. It can be demonstrated theoretically that if Euclidean or Mahalanobis type of distances in the feature space are considered, the data density reduces to Cauchy type as referred in equation (5). It can also be demonstrated that the so called typicality, τ , which is the weighted average of the data density, D, with weights representing the frequency of occurrence of a data sample [6].…”

Section: Concept and Basic Algorithmmentioning

confidence: 99%

“…Density per feature f is obtained according to the equation (5), where D f i denotes the density for f -th feature of thex i sample.…”

Section: Concept and Basic Algorithmmentioning

confidence: 99%

Detecting and Learning from Unknown by Extremely Weak Supervision: eXploratory Classifier (xClass)

2021

Preprint

In this paper, we break with the traditional approach to classification, which is regarded as a form of supervised learning. We offer a method and algorithm, which make possible fully autonomous (unsupervised) detection of new classes, and learning following a very parsimonious training priming (few labeled data samples only). Moreover, new unknown classes may appear at a later stage and the proposed xClass method and algorithm are able to successfully discover this and learn from the data autonomously. Furthermore, the features (inputs to the classifier) are automatically sub-selected by the algorithm based on the accumulated data density per feature per class. In addition, the automatically generated model is easy to interpret, is locally generative and based on prototypes which define the modes of the data distribution. As a result, a highly efficient, lean, human-understandable, autonomously self-learning model (which only needs an extremely parsimonious priming) emerges from the data. To validate our proposal we approbated it on four challenging problems, including imbalanced Faces-1999 data base, Caltech-101 data set, vehicles dataset, and iRoads dataset, which is a dataset of images of autonomous driving scenarios. Not only we achieved higher precision (in one of the problems outperforming by 25% all other methods), but, more significantly, we only used a single class beforehand, while other methods used all the available classes) and we generated interpretable models with smaller number of features used, through extremely weak and weak supervision. We demonstrated the ability to detect and learn new classes for both, images and numerical examples.

“…The reason it is Cauchy is not arbitrary [4]. It can be demonstrated theoretically that if Euclidean or Mahalanobis type of distances in the feature space are considered, the data density reduces to Cauchy type as referred in equation (5). It can also be demonstrated that the so-called typicality, s, which is the weighted average of the data density, D, with weights representing the frequency of occurrence of a data sample [6].…”

Section: Dðmentioning

confidence: 99%

“…So-called reinforcement learning offers some departure from complete labeling, but still requires user input for each individual data sample. The most powerful approaches such as deep learning and support vector machines (SVM) suffer from lack of interpretability [5,11,25,30], are extremely power, time and computational resources hungry and are like dinosaurs-unable to adapt and change with agility. They require complete retraining even for a single or few new data samples.…”

Section: Introductionmentioning

confidence: 99%

Detecting and learning from unknown by extremely weak supervision: exploratory classifier (xClass)

2021

Neural Comput & Applic

In this paper, we break with the traditional approach to classification, which is regarded as a form of supervised learning. We offer a method and algorithm, which make possible fully autonomous (unsupervised) detection of new classes, and learning following a very parsimonious training priming (few labeled data samples only). Moreover, new unknown classes may appear at a later stage and the proposed xClass method and algorithm are able to successfully discover this and learn from the data autonomously. Furthermore, the features (inputs to the classifier) are automatically sub-selected by the algorithm based on the accumulated data density per feature per class. In addition, the automatically generated model is easy to interpret and is locally generative and based on prototypes which define the modes of the data distribution. As a result, a highly efficient, lean, human-understandable, autonomously self-learning model (which only needs an extremely parsimonious priming) emerges from the data. To validate our proposal, we approbated it on four challenging problems, including imbalanced Faces-1999 data base, Caltech-101 dataset, vehicles dataset, and iRoads dataset, which is a dataset of images of autonomous driving scenarios. Not only we achieved higher precision (in one of the problems outperforming by 25% all other methods), but, more significantly, we only used a single class beforehand, while other methods used all the available classes and we generated interpretable models with smaller number of features used, through extremely weak and weak supervision. We demonstrated the ability to detect and learn new classes for both images and numerical examples.