The ASPLOS'18 program is the result of a thorough evaluation process, which we started by forming the program committee (PC) with 50 members and the external review committee (ERC) with 73 members. Moreover, we split up the PC into two independent sub-PCs while keeping the ERC as a single unit. We carefully assigned the PC members to the two groups, ensuring that (1) each sub-PC would cover all ASPLOS topics and (2) the experts on each topic would be evenly split across the sub-PCs. In response to the call for papers, we received 319 submissions, just one shy of last year's record. (This number includes 18 submissions that were either withdrawn by their authors or desk-rejected for clear violations of the formatting rules.) After receiving reviewing bids from most committee members, we also split the submissions evenly across the two sub-PCs, so that each submission would receive reviews from only one sub-PC. We manually moved submissions across sub-PCs to maximize reviewer expertise, according to the PC members' bids. The review process proceeded in two rounds, followed by an extensive online discussion period. During the first round, all submissions received 3-4 reviews. Based on these reviews, we selected 158 submissions to go through the second round of reviews, which produced 2-4 additional reviews for these submissions. During the review process, we also requested reviews from 52 external experts on a case-by-case basis. Throughout the process, our main goal in assigning reviewers to submissions was to maximize reviewer expertise. Overall, the committees and external experts produced 1,227 reviews. After the online discussion period involving all reviewers, we selected 100 submissions for discussion (15 papers online-tagged tentative-accepts and 85 papers online-tagged discuss-at-meeting) during the PC meeting on November 10, 2017 at Georgia Tech. 47 of the 50 PC members were physically present at the meeting and 2 others participated remotely. The whole committee met together in the morning, and split up into the two independent sub- PCs in the afternoon. We did not set a limit for the number of accepted submissions. During the meeting, we accepted 47 submissions and conditionally accepted (subject to shepherding) 9 others. After carefully addressing the reviewers' comments, all shepherded submissions were ultimately accepted. The acceptance rate of the two sub-PCs was exactly the same: 28/151 and 28/150. To complete the program, we invited two outstanding keynote speakers: Hillery Hunter (IBM Research) and Fred Chong (University of Chicago). We are pleased that the 56 accepted submissions and 2 keynote talks represent an exciting spectrum of traditional and emerging ASPLOS topics.
Bridge the Gap between Neural Networks and Neuromorphic Hardware with a Neural Network Compiler
Different from developing neural networks (NNs) for general-purpose processors, the development for NN chips usually faces with some hardware-specific restrictions, such as limited precision of network signals and parameters, constrained computation ...
MAERI: Enabling Flexible Dataflow Mapping over DNN Accelerators via Reconfigurable Interconnects
Deep neural networks (DNN) have demonstrated highly promising results across computer vision and speech recognition, and are becoming foundational for ubiquitous AI. The computational complexity of these algorithms and a need for high energy-efficiency ...
VIBNN: Hardware Acceleration of Bayesian Neural Networks
Bayesian Neural Networks (BNNs) have been proposed to address the problem of model uncertainty in training and inference. By introducing weights associated with conditioned probability distributions, BNNs are capable of resolving the overfitting issue ...
Cited By
- Chen Y, Pei L and Carlson T AfterImage: Leaking Control Flow Data and Tracking Load Operations via the Hardware Prefetcher Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2, (16-32)
-
Sarker M, Saad M, Riaz A, Lipu M and Olazagoitia J (2022). Micro Energy Storage Systems in Energy Harvesting Applications: Analytical Evaluation towards Future Research Improvement, Micromachines, 10.3390/mi13040512, 13:4, (512)
Recommendations
Acceptance Rates
Year | Submitted | Accepted | Rate |
---|---|---|---|
ASPLOS '19 | 351 | 74 | 21% |
ASPLOS '18 | 319 | 56 | 18% |
ASPLOS '17 | 320 | 53 | 17% |
ASPLOS '16 | 232 | 53 | 23% |
ASPLOS '15 | 287 | 48 | 17% |
ASPLOS '14 | 217 | 49 | 23% |
ASPLOS XV | 181 | 32 | 18% |
ASPLOS XIII | 127 | 31 | 24% |
ASPLOS XII | 158 | 38 | 24% |
ASPLOS X | 175 | 24 | 14% |
ASPLOS IX | 114 | 24 | 21% |
ASPLOS VIII | 123 | 28 | 23% |
ASPLOS VII | 109 | 25 | 23% |
Overall | 2,713 | 535 | 20% |