Search papers, labs, and topics across Lattice.
This paper introduces ApproxHDC, a framework that automates the identification and application of domain-specific approximations in Hyperdimensional Computing (HDC) workloads, enhancing performance across various hardware platforms. By leveraging the HPVM-HDC compiler infrastructure, ApproxHDC enables retargetable compilation for CPUs, GPUs, and emerging in-memory computing technologies, such as ReRAM and PCM. The framework effectively navigates the vast space of potential approximations, achieving significant performance improvements with minimal accuracy loss, showcasing HDC's potential as a viable alternative to traditional deep learning methods.
ApproxHDC reveals that automated approximation tuning in Hyperdimensional Computing can yield significant performance boosts without sacrificing accuracy, challenging conventional deep learning paradigms.
As Moore's law reaches its physical and economic limits, domain-specific approaches are increasingly employed to accelerate machine learning workloads. Hyperdimensional Computing (HDC) represents one such emerging paradigm, offering an alternative to conventional deep learning techniques. Rooted in cognitive models of computation, HDC is designed bottom-up with hardware efficiency as a first-class objective. HDC workloads map naturally to heterogeneous hardware platforms, including CPUs, GPUs, and FPGAs, as well as emerging in-memory computing technologies such as Resistive RAM (ReRAM) and Phase-Change Memory (PCM). HDC algorithms are intrinsically tolerant to noise and approximation, enabling substantial performance gains with minimal accuracy loss. In this work, we introduce ApproxHDC, a framework for automated identification and application of domain-specific approximations in HDC workloads. ApproxHDC extends the HPVM-HDC compiler infrastructure to enable retargetable compilation across diverse hardware backends, including CPUs, GPUs, and simulated ReRAM and PCM-based accelerators. The space of possible approximations is exponentially large; ApproxHDC employs efficient search and analysis to navigate it and identify high-impact configurations spanning both software and hardware levels.
