AutoSmart: An Efficient and Automatic Machine Learning Framework for Temporal Relational Data

Abstract

Temporal relational data, perhaps the most commonly used data type in industrial machine learning applications, needs labor-intensive feature engineering and data analyzing for giving precise model predictions. An automatic machine learning framework is needed to ease the manual efforts in fine-tuning the models so that the experts can focus more on other problems that really need humans' engagement such as problem definition, deployment, and business services. However, there are three main challenges for building automatic solutions for temporal relational data: 1) how to effectively and automatically mining useful information from the multiple tables and the relations from them? 2) how to be self-adjustable to control the time and memory consumption within a certain budget? and 3) how to give generic solutions to a wide range of tasks? In this work, we propose our solution that successfully addresses the above issues in an end-to-end automatic way. The proposed framework, AutoSmart, is the winning solution to the KDD Cup 2019 of the AutoML Track, which is one of the largest AutoML competition to date (860 teams with around 4,955 submissions). The framework includes automatic data processing, table merging, feature engineering, and model tuning, with a time&memory controller for efficiently and automatically formulating the models. The proposed framework outperforms the baseline solution significantly on several datasets in various domains. The source code is available at https://github.com/DeepBlueAI/AutoSmart. CCS CONCEPTS• Computing methodologies → Model development and analysis; Machine learning.

From the Beginning: Key Transitions in the First 15 Years of DNSSEC

“…Going forward, we intend to examine the specific implications of popular key transitions, to understand where deployed innovations can advise security standards. To better evaluate and understand noncompliant transitions, it is also possible to make use of statistical approaches common in artificial intelligence and machine learning to cluster similar transitions [42] or even to engineer and select new features beside the ten introduced in this work. Further, we intend to investigate the applicability of using our anatomy for other large-scale object security systems, such as the Resource Public Key Infrastructure (RPKI) [8] and the Web PKI [48].…”

“…In the context of Web PKI, Dong, Kane, and Camp [41] define a set of features to describe X.509 and apply deep neural networks to detect rogue certificates. And finally, in general context of relational data, Lou et al [42] proposes a method to cluster related data.…”

Data augmentation: A comprehensive survey of modern approaches

“…AutoSmart [312] (Figure 25) is another example of a fully automated machine learning framework that performs Figure 24. AutoPrognosis [311] uses patient data to perform end-to-end automated data processing.…”

“…Visualization utilities typically provide all necessary information needed to understand and make useful decisions about the task in relation to the available data. Such capabilities facilitate [312]. The approach performs several data preprocessing and feature engineering tasks in an end-to-end manner.…”

“…Figure25. The general architecture of AutoSmart[312]. The approach performs several data preprocessing and feature engineering tasks in an end-to-end manner.…”