Explainable AI (XAI) and interpretable machine learning methods help to build trust in model predictions and derived insights, yet also present a perverse incentive for analysts to manipulate XAI metrics to support pre-specified conclusions. This paper introduces the concept of X-hacking, a form of p-hacking applied to XAI metrics such as Shap values. We show how easily an automated machine learning pipeline can be adapted to exploit model multiplicity at scale: searching a set of defensible’ models with similar predictive performance to find a desired explanation. We formulate the trade-off between explanation and accuracy as a multi-objective optimisation problem, and illustrate empirically on familiar real-world datasets that, on average, Bayesian optimisation accelerates X-hacking 3-fold for features susceptible to it, versus random sampling. We show the vulnerability of a dataset to X-hacking can be determined by information redundancy among features. Finally, we suggest possible methods for detection and prevention, and discuss ethical implications for the credibility and reproducibility of XAI.

Abstract When designing data science (DS) pipelines, end-users can get overwhelmed by the large and growing set of available data preprocessing and modeling techniques. Intelligent discovery assistants (IDAs) and automated machine learning (AutoML) solutions aim to facilitate end-users by (semi-)automating the process. However, they are expensive to compute and yield limited applicability for a wide range of real-world use cases and application domains. This is due to (a) their need to execute thousands of pipelines to get the optimal one, (b) their limited support of DS tasks, e.g., supervised classification or regression only, and a small, static set of available data preprocessing and ML algorithms; and (c) their restriction to quantifiable evaluation processes and metrics, e.g., tenfold cross-validation using the ROC AUC score for classification. To overcome these limitations, we propose a human-in-the-loop approach for theassisteddesignofdatasciencepipelinesusing previously executed pipelines. Based on a user query, i.e., data and a DS task, our framework outputs a ranked list of pipeline candidates from which the user can choose to execute or modify in real time. To recommend pipelines, it first identifies relevant datasets and pipelines utilizing efficient similarity search. It then ranks the candidate pipelines using multi-objective sorting and takes user interactions into account to improve suggestions over time. In our experimental evaluation, the proposed framework significantly outperforms the state-of-the-art IDA tool and achieves similar predictive performance with state-of-the-art long-running AutoML solutions while being real-time, generic to any evaluation processes and DS tasks, and extensible to new operators.

Existing automated machine learning solutions and intelligent discovery assistants are popular tools that facilitate the end-user with the design of data science (DS) pipelines. However, they yield limited applicability for a wide range of real-world use cases and application domains due to (a) the limited support of DS tasks; (b) a small, static set of available operators; and (c) restriction to evaluation processes with quantifiable loss functions. We demonstrate DORIAN, a human-in-the-loop approach for the assisted design of data science pipelines that supports a large and growing set of DS tasks, operators, and arbitrary user-defined evaluation processes. Based on the user query, i.e., a dataset and a DS task, DORIAN computes a ranked list of candidate pipelines that the end-user can choose from, alter, execute and evaluate. It stores executed pipelines in an experiment database and utilizes similarity-based search to identify relevant previously-run pipelines from the experiment database. DORIAN also takes user interaction into account to improve suggestions over time. We show how users can interact with DORIAN to create and compare DS pipelines on various real-world DS tasks without the need for writing any code.

Data science workflows are largely exploratory, dealing with under-specified objectives, open-ended problems, and unknown business value. Therefore, little investment is made in systematic acquisition, integration, and pre-processing of data. This lack of infrastructure results in redundant manual effort and computation. Furthermore, central data consolidation is not always technically or economically desirable or even feasible (e.g., due to privacy, and/or data ownership). The ExDRa system aims to provide system infrastructure for this exploratory data science process on federated and heterogeneous, raw data sources. Technical focus areas include (1) ad-hoc and federated data integration on raw data, (2) data organization and reuse of intermediates, and (3) optimization of the data science lifecycle, under awareness of partially accessible data. In this paper, we describe use cases, the overall system architecture, selected features of SystemDS' new federated backend (for federated linear algebra programs, federated parameter servers, and federated data preparation), as well as promising initial results. Beyond existing work on federated learning, ExDRa focuses on enterprise federated ML and related data pre-processing challenges. In this context, federated ML has the potential to create a more fine-grained spectrum of data ownership and thus, even new markets.

Reproducibility plays a crucial role in experimentation. However, the modern research ecosystem and the underlying frameworks are constantly evolving and thereby making it extremely difficult to reliably reproduce scientific artifacts such as data, algorithms, trained models and visualizations. We, therefore, aim to design a novel system for assisting data scientists with rigorous end-to-end documentation of data-oriented experiments. Capturing data lineage, metadata, and other artifacts helps reproducing and sharing experimental results. We summarize this challenge as automated documentation of data science experiments. We aim at reducing manual overhead for experimenting researchers, and intend to create a novel approach in dataflow and metadata tracking based on the analysis of the experiment source code. The envisioned system will accelerate the research process in general, and enable capturing fine-grained meta information by deriving a declarative representation of data science experiments.

This paper presents an ongoing study on affective human-robot interaction. In our previous research, touch type is shown to be informative for communicated emotion. Here, a soft matrix array sensor is used to capture the tactile interaction between human and robot and 6 machine learning methods including CNN, RNN and C3D are implemented to classify different touch types, constituting a pre-stage to recognizing emotional tactile interaction. Results show an average recognition rate of 95% by C3D for classified touch types, which provide stable classification results for developing social touch technology.