End-to-End Machine Learning with Apache AsterixDB

Alkowaileet, Wail Y.; Alsubaiee, Sattam; Carey, Michael J.; Li, Chen; Ramampiaro, Heri; Sinthong, Phanwadee; Wang, Xikui

doi:10.1145/3209889.3209894

Cited by 11 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Structured data can be in the form of raw tables as well as streams that users access when curating features. To facilitate sharing of features across an organization and maintaining features if they get updated, feature stores allow for feature authoring and publishing [4]. Users provide simple definitional metadata, e.g., the feature update cadence and a definition SQL query, and upload the definition to the FS.…”

Section: Training Datamentioning

confidence: 99%

Managing ML Pipelines: Feature Stores and the Coming Wave of Embedding Ecosystems

Orr,

Sanyal,

Ling

et al. 2021

Preprint

View full text Add to dashboard Cite

The industrial machine learning pipeline requires iterating on model features, training and deploying models, and monitoring deployed models at scale. Feature stores were developed to manage and standardize the engineer's workflow in this end-to-end pipeline, focusing on traditional tabular feature data. In recent years, however, model development has shifted towards using self-supervised pretrained embeddings as model features. Managing these embeddings and the downstream systems that use them introduces new challenges with respect to managing embedding training data, measuring embedding quality, and monitoring downstream models that use embeddings. These challenges are largely unaddressed in standard feature stores. Our goal in this tutorial is to introduce the feature store system and discuss the challenges and current solutions to managing these new embedding-centric pipelines.

show abstract

Section: Training Datamentioning

confidence: 99%

Managing ML Pipelines: Feature Stores and the Coming Wave of Embedding Ecosystems

Orr,

Sanyal,

Ling

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Traditional solutions aimed at distributed query evaluation and data analytics do not take into consideration access restrictions (e.g., [2,4,15,17,20]). Solutions aimed at enforcing access restrictions in the relational database scenario (e.g., view based access control [8,13,21], access patterns [3,6], data masking [16]) instead do not consider encryption as a solution for protecting confidentiality.…”

Section: Related Workmentioning

confidence: 99%

Distributed Query Evaluation over Encrypted Data

Vimercati

Foresti

Jajodia

et al. 2021

Data and Applications Security and Privacy XXXV

View full text Add to dashboard Cite

The availability of a multitude of data sources has naturally increased the need for subjects to collaborate for distributed computations, aimed at combining different data collections for their elaboration and analysis. Due to the quick pace at which collected data grow, often the authorities collecting and owning such datasets resort to external third parties (e.g., cloud providers) for their storage and management. Data under the control of different authorities are autonomously encrypted (using a different encryption scheme and key) for their external storage. This makes distributed computations combining these sources hard. In this paper, we propose an approach enabling collaborative computations over data encrypted in storage, selectively involving also subjects that might not be authorized for accessing the data in plaintext when it is considered economically convenient.

show abstract

“…The problem of managing queries in distributed scenarios has been extensively studied, but traditional solutions (e.g., [19,21]) as well as modern approaches that consider big data analytics (e.g., [2,4,25]) do not take into consideration access restrictions. In the relational database context, access restrictions can be supported by views (e.g., [9,17,26]), access patterns (e.g., [3,6]), or data masking (e.g., [20]).…”

Section: Related Workmentioning

confidence: 99%

An authorization model for query execution in the cloud

et al. 2021

View full text Add to dashboard Cite

We present a novel approach for the specification and enforcement of authorizations that enables controlled data sharing for collaborative queries in the cloud. Data authorities can establish authorizations regulating access to their data distinguishing three visibility levels (no visibility, encrypted visibility, and plaintext visibility). Authorizations are enforced accounting for the information content carried in the computation to ensure no information is improperly leaked and adjusting visibility of data on-the-fly. Assignment of operations to subjects takes into consideration the cost of operation execution as well as of the encryption/decryption operations needed to make the assignment authorized. Our approach enables users and data authorities to fully enjoy the benefits and economic savings of the competitive open cloud market, while maintaining control over data.

show abstract

End-to-End Machine Learning with Apache AsterixDB

Cited by 11 publications

References 11 publications

Managing ML Pipelines: Feature Stores and the Coming Wave of Embedding Ecosystems

Managing ML Pipelines: Feature Stores and the Coming Wave of Embedding Ecosystems

Distributed Query Evaluation over Encrypted Data

An authorization model for query execution in the cloud

Contact Info

Product

Resources

About