Skip to yearly menu bar Skip to main content

Tianyi Chen · Pin-Yu Chen · Nathalie Baracaldo Angel · Carlee Joe-Wong

[ Mission Ballroom MR2 ]

Federated Learning (FL) has recently emerged as the overarching framework for distributed machine learning (ML) beyond data centers. FL, in both cross-device and cross-silo, enables collaborative ML model training from originally isolated data without sacrificing data privacy. Such potential use of FL has since then attracted an explosive attention from the ML, computer systems, optimization, signal processing, wireless networking, data mining, computer architecture, privacy and security communities.

FL-related research is penetrating into almost every science and engineering discipline. However, as FL comes closer to being deployable in real-world systems, many open problems in FL today cannot be solved solely by researchers in one community. For example, designing most effcient and reliable FL algorithms require leveraging expertises from systems, security, signal processing and networking communities. On the other hand, designing most effcient and scalable computing and networking systems require leveraging collaborative advances from ML, data mining, and optimization communities.

In light of the differences in education backgrounds, toolboxes, viewpoints, and design principles of different communities, this workshop aims to break community barriers and bring researchers from pertinent communities together to address open problems in FL. More importantly, this workshop aims to stimulate discussion among experts in different fields (e.g., industry and …

Tom St John · Murali Emani

[ Mission Ballroom B4 ]

With evolving system architectures, hardware and software stack, diverse machine learning workloads, and data, it is important to understand how these components interact with each other. Well-defined benchmarking procedures help evaluate and reason the performance gains with ML workload to system mapping.

Key problems that we seek to address are: (i) which representative ML benchmarks cater to workloads seen in industry, national labs, and interdisciplinary sciences; (ii) how to characterize the ML workloads based on their interaction with hardware; (iii) what novel aspects of hardware, such as heterogeneity in compute, memory, and bandwidth, will drive their adoption; (iv) performance modeling and projections to next-generation hardware.

The workshop will invite experts in these research areas to present recent work and potential directions to pursue. Accepted papers from a rigorous evaluation process will present state-of-the-art research efforts. We have also secured funding from MLCommonsTM to provide a best paper award for an outstanding submission. A panel discussion will foster an interactive platform for discussion between speakers and the audience.

Jian Zhang · Christina Delimitrou · Qingwei Lin · Daniel Crankshaw

[ Mission Ballroom B5 ]

Summary: We propose a full-day workshop (2 keynote, 3 invited talks, 6 accepted papers, 1 panel, and 1 poster session) at MLSys’22 conference for professionals, researchers, and practitioners who are interested in leveraging artificial intelligence and machine learning to efficiently design and build
cloud computing systems and operating cloud services. This workshop will be guided by the steering committee and the program committee with members from both academia and industry in systems, AI/ML, and software engineering areas.

Deniz Altınbüken · Lyric Doshi · Milad Hashemi · Martin Maas

[ Mission Ballroom MR1 ]

Using ML for improving computer systems has seen a significant amount of work both in academia and industry. However, deployed uses of such techniques remain rare. While many published works in this space focus on solving the underlying learning problems, we observed from an industry vantage point that some of the biggest challenges of deploying ML for Systems in practice come from non-ML systems aspects, such as feature stability, reliability, availability, ML integration into rollout processes, verification, safety guarantees, feedback loops introduced by learning, debuggability, and explainability.
The goal of this workshop is to raise awareness of these problems and bring together practitioners (both on the production systems and ML side) and academic researchers, to work towards a methodology of capturing these problems in academic research. We believe that starting this conversation between the academic and industrial research communities will facilitate the adoption of ML for Systems research in production systems, and will provide the academic community with access to new research problems that exist in real-world deployments but have seen less attention in the academic community.
The workshop will uniquely facilitate this conversation by providing a venue for lightweight sharing of anecdotes and experiences from real-world deployments, as well …