initialized by constructor call can be const and may also be easier to use with standard containers or algorithms. The problems with doing work in constructors are: • If the work calls virtual functions, these Lambdas are much more concise than other ways of defining function objects to be passed to STL algorithms, which can be a readability im- provement. • Appropriate use of default captures can remove redundancy the error messages are understandable and actionable from a user point of view. Boost Use only approved libraries from the Boost library collection. The Boost library collection is a popular collection

Archival publications, awareness of peer-review  From right communities (e.g., ML systems vs ML algorithms)  Reputation of website, authors, etc  Recap: Give Credit  Cite broadly, give credit to inspiring … \end{verbatim}  Theorem, Definition, Examples  Refine theorem environments as needed  Algorithms  Can be clearer than text, but not always  Carefully select the right level of abstraction Avoid exclamation marks! Never use more than one!! Scientific Writing When using disk[,] tree algorithms were found to be particularly poor. We use index structures like b-trees, tries, and hash tables;

selection needs time  pipeline model  Ex. Compressed Linear Algebra  Problem: Iterative ML algorithms + memory-bandwidth-bound operations  crucial to fit data in memory  automatic lossless compression of Technology, WS 2020/21 Abstract % 1. State the problem Large-scale machine learning (ML) algorithms are often iterative, using repeated read-only data access and I/O-bound matrix-vector multiplications