Abstract
Concerns with the interpretability of ML models are growing as the technology is used in increasingly sensitive domains (e.g., health and public administration). Synthetic data can be used to understand models better, for instance, if the examples are generated close to the frontier between classes. However, data augmentation techniques, such as Generative Adversarial Networks (GAN), have been mostly used to generate training data that leads to better models. We propose a variation of GANs that, given a model, generates realistic data that is classified with low confidence by a given classifier. The generated examples can be used in order to gain insights on the frontier between classes. We empirically evaluate our approach on two well-known image classification benchmark datasets, MNIST and Fashion MNIST. Results show that the approach is able to generate images that are closer to the frontier when compared to the original ones, but still realistic. Manual inspection confirms that some of those images are confusing even for humans.

Abstract
The amount of data produced by distributed devices, such as smart devices and the IoT, is increasing continuously. The cost of transmitting data and also distributed computing power raise interest in distributed data mining (DDM). However, in a pure DDM scenario, data availability may not be enough to generate reliable models in a distributed environment. So, the ability to exchange data efficiently and effectively will become a crucial component of DDM. In this paper, we propose the concept of the Machine Learning Data Market (MLDM), a framework for the exchange of data among autonomous agents. We consider a set of learning agents in a cooperative distributed ML, where agents negotiate data to improve the models they use locally. In the proposed data market, the system's predictive accuracy is investigated, as well as the economic value of data. The question addressed in this paper is: How data exchange among the agents will improve the accuracy of the learning model. Agent budget is defined as a limitation of negotiation. We defined a multi-agent system with negotiation and assessed it against the multi-agent system baseline and the single-agent system. The proposed framework is analyzed based on the different sizes of batch data collected over time to find out how this changes the effect of the negotiation on the accuracy of the model. The results indicate that even simple negotiation among agents increases their learning accuracy.

Abstract
Several meta-learning approaches have been developed for the problem of algorithm selection. In this context, it is of central importance to collect a sufficient number of datasets to be used as meta-examples in order to provide reliable results. Recently, some proposals to generate datasets have addressed this issue with successful results. These proposals include datasetoids, which is a simple manipulation method to obtain new datasets from existing ones. However, the increase in the number of datasets raises another issue: in order to generate meta-examples for training, it is necessary to estimate the performance of the algorithms on the datasets. This typically requires running all candidate algorithms on all datasets, which is computationally very expensive. One approach to address this problem is the use of active learning, termed active meta-learning. In this paper we investigate the combined use of active meta-learning and datasetoids. Our results show that it is possible to significantly reduce the computational cost of generating meta-examples not only without loss of meta-learning accuracy but with potential gains.

Abstract

Abstract
We present a meta-learning method to support selection of candidate learning algorithms. It uses a k-Nearest Neighbor algorithm to identify the datasets that are most similar to the one at hand. The distance between datasets is assessed using a relatively small set of data characteristics, which was selected to represent properties that affect algorithm performance. The performance of the candidate algorithms on those datasets is used to generate a recommendation to the user in the form of a ranking. The performance is assessed using a multicriteria evaluation measure that takes not only accuracy, but also time into account. As it is not common in Machine Learning to work with rankings, we had to identify and adapt existing statistical techniques to devise an appropriate evaluation methodology. Using that methodology, we show that the meta-learning method presented leads to significantly better rankings than the baseline ranking method. The evaluation methodology is general and can be adapted to other ranking problems. Although here we have concentrated on ranking classification algorithms, the meta-learning framework presented can provide assistance in the selection of combinations of methods or more complex problem solving strategies.

Abstract