Outline ● What is plugin architecture? ● Why use plugin architecture? ● Designing a simplified plugin architecture ● Library used in robotics to implement plugin based system ○ Pluginlib ● Case study study for plugin architecture - MoveIt ● Limitations ● Summary 2Introduction •Abi Sivaraman •Robotics Engineer at PickNik Robotics •I work with robotic arms •MoveIt Maintainer 3What is plugin plugin architecture? Software Design Pattern that allows for developers to add functionality to a larger system without having to alter the source code of the system itself. Plug-ins are self-contained

API server-side architecture for Beginners GopherCon ���� ����.��.�� - @hgsgtk © ����-���� BASE, Inc. � Talk abstract • A practical approach to build server-side architecture in a Go project � Problem of building architecture for beginners � Approach to build architecture � Summary � Talk structure © ����-���� BASE, Inc. � Problem of building architecture for beginners � Approach Approach to build architecture � Summary � Talk structure © ����-���� BASE, Inc. � Why I need server-side architecture �.Keep a design easy to change • -> Separate external input/output and business

uptake in many different computing sectors. The book also contains many insights about computer architecture in general, as well as explaining the particular de- sign choices we made in creating RISC-V. the point, and complete. The book’s commentaries provide a gratuitous history, motivation, and architecture critique. —C. Gordon Bell, Microsoft and designer of the Digital PDP-11 and VAX-11 instruction handy little book effortlessly summarizes all the essential elements of the RISC-V Instruction Set Architecture, a perfect reference guide for students and practitioners alike. —Professor Randy Katz, University

embedded software • Started using C++ in 1995 • First cross-platform project in 1994Cross-Platform Architecture Goals • Take advantage of all platforms • Focus on the compiler • Minimize boilerplate and unnecessary requiring implementations that differ depending upon the target machine architecture. • Features may be hardware: CPU architecture, SIMD instruction set, DMA controller, GPIO module, etc. • Features

Feature: binaryBusinessTime; ReadKV 是一个 IO 类型的 OP ReadKV('mtptpoionlinefeatureexp','_id',_id,'ba_search.platform_poi_ business_hour_new.binarybusinesstime','STRING') // FeatureA : CtxDateInfo; ParseJSON 和多目标相关的工作，欢迎业界同行一起交流。 参考资料 [1] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]//Advances in neural information processing systems. 2017: 5998-6008. [2] Devlin J, Chang M W, Lee K, et al. Bert: Pre-training Song W, Shi C, Xiao Z, et al. Autoint: Automatic feature interaction learning via self-attentive neural networks[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge

Optimal Transport Assignment for Object Detection, https://arxiv.org/ abs/2103.14259 [8] Computer Architecture: A Quantitative Approach [9] SIoU Loss: More Powerful Learning for Bounding Box Regression, https:// CIKM 论文：Trilateral Spatiotemporal Attention Network for User Behavior Modeling in Location-based Search[23]。 算法 < 61 图 19 基于三边时空注意力机制的用户行为序列网络 在实际建模中，相对于比赛涉及到更多线上部分，而比赛主要专注于离线数据集的精 度极值。同 Debiasing and Yue Hu. 2020. Graph neural architecture search. In IJCAI, Vol. 20. 1403–1409. [12] Matheus Nunes and Gisele L Pappa. 2020. Neural Architecture Search in Graph Neural Networks. In Brazilian Conference

and then use that knowledge to make predictions ◮ Applications: ◮ Computer vision ◮ Medicine ◮ Search engines 6 / 50Embedded devices ◮ Computing devices designed to perform specific tasks within larger paradigm in which an algorithm learns from labeled data to make predictions 11 / 50Neural network (NN) 13 / 50Convolutional neural network (CNN) ◮ Efficient in image classification ◮ A convolutional layer can 50Software dependencies ◮ TensorFlow Lite ◮ OpenCV 16 / 50MobileNet pre-trained model ◮ CNN architecture created by Google ◮ Trained on 1000 classes ◮ Accepts 224x224-pixel images, with 3 color channels

etc. 4 Changkun Ou. 2023. Go on GPU. GopherChina 2023. Session "Foundational Toolchains" CPU Architecture 5 RAM Exec Unit L1 Cache Exec Unit L1 Cache Exec Unit L1 Cache Exec Unit Cache L2 Cache Controller … Changkun Ou. 2023. Go on GPU. GopherChina 2023. Session "Foundational Toolchains" GPU Architecture Memory SIMD Exec Unit Cache SIMD Exec Unit Cache SIMD Exec Unit Cache SIMD Exec Unit also a classic performance improvement problem to solve. Example: Feedforward propagation in neural network is done via matrix multiplication; many other linear solvers rely on matrices, etc 4x5 5x2

What we do at Snap with C++ 2 Neural style transfer Face tracking Full body tracking Cloth simulation Ray tracing Wrist trackingThank you, Serhii Huralnik and Eduardo Madrid!! 3Section 0. Introduction V ABI Procedure Call Standard for the Arm® 64-bit Architecture armv8-a clang 18.1.0 -O2 -std=c++20 x86-64 (AMD64) System V ABI AMD64 Architecture Processor Supplement x86-64 gcc 14.2 -O2 -std=c++20 System V ABI Procedure Call Standard for the Arm® Architecture armv7-a clang 11.0.1 -O2 -std=c++20 x86 (IA-32) System V ABI Intel386 Architecture Processor Supplement x86-64 gcc 14.2 -O2 -std=c++20

on that data, use it to make a model that will hopefully apply in all cases. The focus will be on neural nets, because that is what is used later on in this paper.12 Neurons A neuron is, put simply, a Pitts made a simple mathematical model of a neuron13, which is used as a basis to make more advanced neural nets. It looks like Figure 1, and calculates output in the following procedure: 11Ibid., p. 243 does not make sense from a biological perspective. 4.2.1 The Perceptron The perceptron is a simple neural net, that is it is a network of neurons. It essentially is a row of neurons, that all get the same