Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper surveys 42 studies to define Algorithm Debt (AD) as a type of technical debt impacting ML/DL system performance and scalability. The review identifies the implicit presence and "smells" of AD within ML/DL systems, providing a structured understanding of this often-overlooked issue. The work proposes future research directions to mitigate AD and improve the reliability of ML/DL deployments.
Algorithm Debt, a hidden drag on ML/DL performance and scalability, gets a formal definition and a "smell test" to help you sniff it out in your own systems.
Abstract. The adoption of Machine and Deep Learning (ML/DL) technologies introduces maintenance challenges, leading to Technical Debt (TD). Algorithm Debt (AD) is a TD type that impacts the performance and scalability of ML/DL systems. A review of 42 primary studies expanded AD’s definition, uncovered its implicit presence, identified its smells, and highlighted future directions. These findings will guide an AD-focused study, enhancing the reliability of ML/DL systems.