Deep Learning in the Wild

Abstract

Deep learning with neural networks is applied by an increasing number of people outside of classic research environments, due to the vast success of the methodology on a wide range of machine perception tasks. While this interest is fueled by beautiful success stories, practical work in deep learning on novel tasks without existing baselines remains challenging. This paper explores the specific challenges arising in the realm of real world tasks, based on case studies from research & development in conjunction with industry, and extracts lessons learned from them. It thus fills a gap between the publication of latest algorithmic and methodical developments, and the usually omitted nitty-gritty of how to make them work. Specifically, we give insight into deep learning projects on face matching, print media monitoring, industrial quality control, music scanning, strategy game playing, and automated machine learning, thereby providing best practices for deep learning in practice.

Automated Machine Learning in Practice: State of the Art and Recent Results

“…In this section, we evaluate the usefulness of AutoML for application in business and industry by empirically comparing the most successful automated machine learning algorithms with (a) an industrial prototype as well as (b) a straight-forward improvement inspired by Hyperband [13], [43] (c). This selection spans a wide range of different approaches for pipeline optimization (see Section III) to tackle the CASH problem: the industrial prototype DSM [44] uses random model and hyperparameter search and thus serves as a baseline; Auto-sklearn [17] has won the recent AutoML challenges [33]. Additionally, we report results with TPOT [45], which is developed based on genetic programming [40] instead of Auto-sklearn's Bayesian optimization.…”

Lessons Learned from Challenging Data Science Case Studies

“…Many of the results there are driven by deep learning technology, and the lines between fundamental and applied research have become reasonably blurred in recent years (with companies producing lots of fundamental results, and universities engaging in many different application areas, compare e.g. Stadelmann et al (2018)). This speaks strongly for collaborations between scientists and engineers from different organisations and units that complement each other's knowledge and skills, e.g.…”

Two to Trust: AutoML for Safe Modelling and Interpretable Deep Learning for Robustness

“…The recent success of machine learning (ML) and deep learning (DL) has triggered enormous interest in practical applications of these algorithms in many organizations [17,18]. The emergence of automated ML (AutoML), which includes automated DL (AutoDL), further expands the horizons of such machine learning applications for non-experts and broadens the feasibility of exploring larger search spaces during development.…”