RTML: Large: Collaborative: Harmonizing Predictive Algorithms and Mixed Signal/Precision Circuits via Computation-Data Access Exchange and Adaptive Dataflows

RTML：大型：协作：通过计算数据访问交换和自适应数据流协调预测算法和混合信号/精密电路

• 批准号: 1937435
• 负责人: Yiran Chen
• 金额: $ 25万
• 依托单位: Duke University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1937435&HistoricalAwards=false
• 关键词: RTML; Large; Collaborative; Harmonizing; Predictive

Recent advances in machine learning are fueling a growing demand for intelligent Internet of Things (IoT), i.e., edge network applications. Many of them, such as autonomous vehicles, robots, and healthcare wearables, require real-time and in-situ learning to be perceived as truly intelligent. However, the limited computing and energy resources available at the edge device (e.g., mobile devices, sensors) stand at odds with the massive and growing cost of state-of-the-art machine learning training, posing a grand challenge for real-time machine learning (RTML) at the edge. This goal of this project is to foster a systematic breakthrough in achieving efficient online training of state-of-the-art machine learning algorithms in pervasive resource-constrained platforms and applications. An order of magnitude advance in RTML would enable numerous edge devices to proactively interpret and learn from new data, improve their own performance using what they have learned, and adapt to dynamic environments, all in real time. Success in this project will enable truly intelligent edge devices to penetrate all walks of life and thus generate significant impacts on societies and economies. This project will lead to new courses and open-education resources that can attract diverse groups of students and eventually deliver a platform for inclusion and innovation. The project addresses the RTML grand challenge using a three-pronged 'co-design' approach that seamlessly integrates algorithm, architecture, and circuit-level innovations. Specifically, at the algorithm level, an efficient training framework for RTML, for which trained models are also natively efficient for inference, will be established. Aggressive time and energy reductions can be achieved, at first by improving general training techniques, and then by focusing particularly on online learning and adaptation. At the architecture level, the project will first target reducing the high cost of data movement by trading it for lower-cost computation, and then generate optimal dataflows and hardware architectures to maximize the joint benefits of algorithms and hardware. At the circuit level, the project will leverage adaptive low-precision algorithms and architectures to design ultra-energy-efficient mixed-signal compute fabrics. Statistical computing techniques will be incorporated to demonstrate efficient, scalable, and robust machine learning chips. Finally, at the system level, an integration effort will be included to aid the realization of realistic system goals and to evaluate the innovations of the three core thrusts.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

机器学习的最新进展正在推动对智能物联网（IoT）的需求不断增长，即，边缘网络应用。其中许多，如自动驾驶汽车，机器人和医疗可穿戴设备，需要实时和原位学习才能被视为真正的智能。然而，在边缘设备处可用的有限计算和能量资源（例如，移动的设备、传感器）与最先进的机器学习训练的巨大且不断增长的成本不一致，这对边缘的实时机器学习（RTML）构成了巨大的挑战。该项目的目标是在普遍的资源受限平台和应用程序中实现最先进的机器学习算法的有效在线训练方面取得系统性突破。RTML的一个数量级的进步将使许多边缘设备能够主动解释和学习新数据，使用它们所学到的知识提高自己的性能，并适应动态环境，所有这些都是真实的时间。该项目的成功将使真正的智能边缘设备渗透到各行各业，从而对社会和经济产生重大影响。该项目将带来新的课程和开放教育资源，吸引不同的学生群体，并最终提供一个包容和创新的平台。 该项目使用三管齐下的“协同设计”方法解决了RTML的重大挑战，该方法无缝集成了算法、架构和电路级创新。具体而言，在算法层面，将建立一个有效的RTML训练框架，训练模型也是天生有效的推理。可以通过改进一般培训技术，然后特别关注在线学习和适应，来大幅减少时间和精力。在架构层面，该项目将首先以降低数据移动的高成本为目标，将其转换为低成本计算，然后生成最佳的硬件架构和硬件架构，以最大限度地提高算法和硬件的联合效益。在电路层面，该项目将利用自适应低精度算法和架构来设计超节能混合信号计算结构。将采用统计计算技术来展示高效、可扩展和强大的机器学习芯片。最后，在系统层面，将包括一个集成的努力，以帮助实现现实的系统目标，并评估三个核心thrusts.This奖项的创新反映了NSF的法定使命，并已被认为是值得的支持，通过评估使用基金会的智力价值和更广泛的影响审查标准。

项目成果

PENNI: Pruned Kernel Sharing for Efficient CNN Inference

• 发表时间: 2020-05
• 期刊: ArXiv
• 作者: Shiyu Li;Edward Hanson;H. Li;Yiran Chen

DyNNamic: Dynamically Reshaping, High Data-Reuse Accelerator for Compact DNNs

• DOI: 10.1109/tc.2022.3184272
• 发表时间: 2023-03
• 期刊: IEEE Transactions on Computers
• 影响因子: 3.7
• 作者: Edward Hanson;Shiyu Li;Xuehai Qian;H. Li;Yiran Chen

NASRec: Weight Sharing Neural Architecture Search for Recommender Systems

NASRec：推荐系统的权重共享神经架构搜索

• DOI: 10.1145/3543507.3583446
• 发表时间: 2023
• 期刊: the ACM Web Conference 2023
• 作者: Zhang, Tunhou;Cheng, Dehua;He, Yuchen;Chen, Zhengxing;Dai, Xiaoliang;Xiong, Liang;Yan, Feng;Li, Hai;Chen, Yiran;Wen, Wei
• 通讯作者: Wen, Wei

PIDS: Joint Point Interaction-Dimension Search for 3D Point Cloud

PIDS：3D 点云的联合点交互维度搜索

• 发表时间: 2023
• 期刊: EEE/CVF Winter Conference on Applications of Computer Vision
• 作者: Zhang, Tunhou;Ma, Mingyuan;Yan, Feng;Li, Hai;Chen, Yiran
• 通讯作者: Chen, Yiran

Cascading structured pruning: enabling high data reuse for sparse DNN accelerators

• DOI: 10.1145/3470496.3527419
• 发表时间: 2022-06
• 期刊: Proceedings of the 49th Annual International Symposium on Computer Architecture
• 作者: Edward Hanson;Shiyu Li;H. Li;Yiran Chen