Mind Mappings: Enabling Efficient Algorithm-Accelerator Mapping Space Search

by   Kartik Hegde, et al.

Modern day computing increasingly relies on specialization to satiate growing performance and efficiency requirements. A core challenge in designing such specialized hardware architectures is how to perform mapping space search, i.e., search for an optimal mapping from algorithm to hardware. Prior work shows that choosing an inefficient mapping can lead to multiplicative-factor efficiency overheads. Additionally, the search space is not only large but also non-convex and non-smooth, precluding advanced search techniques. As a result, previous works are forced to implement mapping space search using expert choices or sub-optimal search heuristics. This work proposes Mind Mappings, a novel gradient-based search method for algorithm-accelerator mapping space search. The key idea is to derive a smooth, differentiable approximation to the otherwise non-smooth, non-convex search space. With a smooth, differentiable approximation, we can leverage efficient gradient-based search algorithms to find high-quality mappings. We extensively compare Mind Mappings to black-box optimization schemes used in prior work. When tasked to find mappings for two important workloads (CNN and MTTKRP), the proposed search finds mappings that achieve an average 1.40×, 1.76×, and 1.29× (when run for a fixed number of steps) and 3.16×, 4.19×, and 2.90× (when run for a fixed amount of time) better energy-delay product (EDP) relative to Simulated Annealing, Genetic Algorithms and Reinforcement Learning, respectively. Meanwhile, Mind Mappings returns mappings with only 5.32× higher EDP than a possibly unachievable theoretical lower-bound, indicating proximity to the global optima.



There are no comments yet.


page 1

page 2

page 3

page 4


Rethinking Co-design of Neural Architectures and Hardware Accelerators

Neural architectures and hardware accelerators have been two driving for...

RHNAS: Realizable Hardware and Neural Architecture Search

The rapidly evolving field of Artificial Intelligence necessitates autom...

Hit-and-Run for Sampling and Planning in Non-Convex Spaces

We propose the Hit-and-Run algorithm for planning and sampling problems ...

Marvel: A Data-centric Compiler for DNN Operators on Spatial Accelerators

The efficiency of a spatial DNN accelerator depends heavily on the compi...

Domain-specific Genetic Algorithm for Multi-tenant DNNAccelerator Scheduling

As Deep Learning continues to drive a variety of applications in datacen...

Mirror-Prox SCA Algorithm for Multicast Beamforming and Antenna Selection

This paper considers the (NP-)hard problem of joint multicast beamformin...

Deep regression on manifolds: a 3D rotation case study

Many problems in machine learning involve regressing outputs that do not...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.