Difference between revisions of "Resource:Seminar"

From MobiNetS
Jump to: navigation, search
 
(10 intermediate revisions by 2 users not shown)
Line 1: Line 1:
{{SemNote
{{SemNote
|time='''2025-12-20 10:30'''
|time='''2026-04-10 10:30'''
|addr=4th Research Building A518
|addr=4th Research Building A518
|note=Useful links: [[Resource:Reading_List|📚 Readling list]]; [[Resource:Seminar_schedules|📆 Schedules]]; [[Resource:Previous_Seminars|🧐 Previous seminars]].
|note=Useful links: [[Resource:Reading_List|📚 Readling list]]; [[Resource:Seminar_schedules|📆 Schedules]]; [[Resource:Previous_Seminars|🧐 Previous seminars]].
Line 8: Line 8:


{{Latest_seminar
{{Latest_seminar
|abstract = Low Earth Orbit (LEO) satellite networks are expected to enable global connectivity for next-generation communications. To provide space-centric solutions, the limited coverage time and limited resources of LEO satellites pose challenges to maintaining service continuity and ensuring low latency for users. Furthermore, LEO satellites rely on solar panels to obtain energy, so a balance needs to be struck between energy consumption and service provision for satellite mobile edge computing. In this paper, we aim to achieve space-centric computational task offloading in LEO satellite networks. The goal is to minimize end-to-end task offloading latency while considering the constraints posed by the limited onboard computing, storage, and energy resources in constantly moving LEO satellites. To achieve this, we formulate a joint problem of service migration and power control in energy-harvesting LEO satellite networks. The problem is then converted into a Markov decision process (MDP) and solved with SpaceEdge, a novel algorithm based on Deep Reinforcement Learning (DRL). SpaceEdge offers supports for both centralized learning and multi-agent learning. Simulation results show that SpaceEdge, particularly the multi-agent model, outperforms existing solutions, demonstrating its effectiveness in deploying space-centric task offloading services in LEO satellite networks.
|abstract = To effectively utilize heterogeneous specialized hardware units in modern GPUs, such as TensorCores and Tensor Memory Accelerators, this paper introduces PipeThreader, a new DNN compiler. PipeThreader proposes shifting scheduling functionality from hardware to software so as to enable more efficient and sophisticated computation pipelining with minimal manual effort. This is achieved through sTask-graph, a new DNN computation abstraction, a hierarchical hardware abstraction that captures the capabilities of specialized units, and new scheduling primitives. As a result, PipeThreader can discover efficient pipeline scheduling for well-studied DNN architectures like FlashAttention, achieving comparable or even superior performance. Additionally, it can uncover novel pipeline schemes for emerging models like Mamba2, delivering significantly better performance compared to state-of-the-art hand-crafted implementations. The code is open-sourced at https://github.com/tile-ai/tilelang.
|confname =TWC'24
|confname =OSDI'25
|link = https://ieeexplore.ieee.org/abstract/document/10623400
|link = https://www.usenix.org/conference/osdi25/presentation/cheng
|title= SpaceEdge: Optimizing Service Latency and Sustainability for Space-Centric Task Offloading in LEO Satellite Networks
|title= PipeThreader: Software-defined pipelining for efficient DNN execution
|speaker=Haifeng
|speaker=Junzhe
|date=2025-12-20
|date=2026-4-9
}}
{{Latest_seminar
|abstract =For highly immersive mobile volumetric video streaming, it is essential to deliver photo-realistic full-scene content with smooth playback. Unlike traditional representations such as point clouds, 3D Gaussian Splatting (3DGS) has gained attention for its ability to represent high-quality full-scene 3D content. However, our preliminary experiments show that existing methods for 3DGS-based videos fail to achieve smooth playback on mobile devices. In this paper, we propose Vega, a 3DGS-based photo-realistic full-scene volumetric video streaming system that ensures real-time playback on mobile devices. The core idea behind Vega's real-time rendering is object-level selective computation, which allocates computational resources to visually important objects to meet strict rendering deadlines. To enable mobile streaming based on the selective computation, Vega addresses two challenges: (1) designing an encoding scheme that optimizes the data size of videos while being compatible with object-level prioritization, and (2) developing a rendering pipeline that efficiently operates on resource-constrained mobile devices. We implemented an end-to-end Vega system, consisting of a streaming server and an Android application. Experimental results on commodity smartphones show that Vega achieves 30 frames per second (FPS) for full-scene volumetric video streaming while maintaining competitive data size and visual quality compared to existing baselines.
|confname =Mobicom'25
|link = https://dl.acm.org/doi/10.1145/3680207.3765267
|title= Vega: Fully Immersive Mobile Volumetric Video Streaming with 3D Gaussian Splatting
|speaker=Jiyi
|date=2025-12-20
}}
}}
{{Resource:Previous_Seminars}}
{{Resource:Previous_Seminars}}

Latest revision as of 10:37, 10 April 2026

Time: 2026-04-10 10:30
Address: 4th Research Building A518
Useful links: 📚 Readling list; 📆 Schedules; 🧐 Previous seminars.

Latest

  1. [OSDI'25] PipeThreader: Software-defined pipelining for efficient DNN execution, Junzhe
    Abstract: To effectively utilize heterogeneous specialized hardware units in modern GPUs, such as TensorCores and Tensor Memory Accelerators, this paper introduces PipeThreader, a new DNN compiler. PipeThreader proposes shifting scheduling functionality from hardware to software so as to enable more efficient and sophisticated computation pipelining with minimal manual effort. This is achieved through sTask-graph, a new DNN computation abstraction, a hierarchical hardware abstraction that captures the capabilities of specialized units, and new scheduling primitives. As a result, PipeThreader can discover efficient pipeline scheduling for well-studied DNN architectures like FlashAttention, achieving comparable or even superior performance. Additionally, it can uncover novel pipeline schemes for emerging models like Mamba2, delivering significantly better performance compared to state-of-the-art hand-crafted implementations. The code is open-sourced at https://github.com/tile-ai/tilelang.

History

2024

2023

2022

2021

2020

  • [Topic] [ The path planning algorithm for multiple mobile edge servers in EdgeGO], Rong Cong, 2020-11-18

2019

2018

2017

Instructions

请使用Latest_seminar和Hist_seminar模板更新本页信息.

    • 修改时间和地点信息
    • 将当前latest seminar部分的code复制到这个页面中
    • 将{{Latest_seminar... 修改为 {{Hist_seminar...，并增加对应的日期信息|date=
    • 填入latest seminar各字段信息
    • link请务必不要留空，如果没有link则填本页地址 https://mobinets.org/index.php?title=Resource:Seminar
  • 格式说明
    • Latest_seminar:

{{Latest_seminar
|confname=
|link=
|title=
|speaker=
}}

    • Hist_seminar

{{Hist_seminar
|confname=
|link=
|title=
|speaker=
|date=
}}

Retrieved from "http://mobinets.cn/site/index.php?title=Resource:Seminar&oldid=3511"