CONQuER: Hardware-Aware Mixed-Precision Quantisation with Online-Calibrated Surrogates

• URL: http://arxiv.org/abs/2607.25884v1
• Date: Tue, 28 Jul 2026 15:45:26 GMT
• Title: CONQuER: Hardware-Aware Mixed-Precision Quantisation with Online-Calibrated Surrogates
• Abstract summary: We present CONQuER, a unified compiler-integrated infrastructure for hardware-aware MPQ.<n> CONQuER shifts quantisation into the compiler pipeline at the TOSA level, enabling intelligent configuration handling.<n>We show that CONQuER discovers configurations up to 12.19x faster with top-1 accuracy within 1.44% of the unquantised baseline.
• Score: 3.4090089330250724
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deploying deep neural networks on resource-constrained hardware relies on mixed-precision quantisation (MPQ). current deployment toolchains severely fragment this process. Quantisation typically occurs as a hardware-agnostic preprocessing step in front-end frameworks, disconnected from the downstream compilers that generate the physical machine code. This separation leads to suboptimal configurations where assigned bit-widths map poorly to the target machine's heterogeneous hardware execution blocks such as tensor cores and variable-width vector units, incurring severe runtime execution penalties. Furthermore, evaluating these configurations via exhaustive hardware-in-the-loop (HIL) testing is intractable due to the exponentially large search space. We present CONQuER, a unified compiler-integrated infrastructure for hardware-aware MPQ. CONQuER shifts quantisation into the compiler pipeline at the TOSA level, enabling intelligent configuration handling based on compiler support. To evaluate this combinatorial search space of different of model layers within practical compilation budgets, CONQuER couples an NSGA-II evolutionary algorithm with a dual-surrogate prescreening engine. This engine evaluates theoretical cache memory bounds and feature space isotropy to discard non-viable configurations. CONQuER then executes only the strongest candidate policies on hardware via IREE, feeding the execution metrics into an online calibrator. This calibrator aligns the surrogate models with the true hardware behaviour during an NSGA-II evolutionary search. Evaluation across mobile and laptop CPUs, and server GPUs demonstrates that optimal quantisation policies are hardware-dependent. By coupling quantisation with compiler lowering and physical execution, CONQuER discovers Pareto-optimal configurations up to 12.19x faster inference with top-1 accuracy within 1.44% of the unquantised baseline.

Related papers

• Quantum Compiler Design for Fault-Tolerant Quantum Computing [14.739903011159207]
  This survey presents a full-stack view of compiler design for QEC-protected quantum computation.<n>We organize existing work into three interacting layers: logical-level QEC compilation, physical-level QEC realization, and runtime integration.
  arXiv  Detail & Related papers  (2026-09-15T17:05:49Z)
• CUDA-Q Logical: Retargetable Compilation for Fault-Tolerant Quantum Computing [30.490749390369032]
  We present Q-Q Logical, a compiler infrastructure for fault-tolerant compilation, analysis, and execution.<n>By deriving every resource estimate directly from compiler artifacts, the framework unifies compilation and resource analysis.<n>We show that schedule estimates reconcile with established independent models.
  arXiv  Detail & Related papers  (2026-09-11T18:00:16Z)
• Heuristic-Based Merging of HPC Traces to Extend Hardware Counter Coverage [0.08399688944263843]
  We propose a methodology to merge execution traces from multiple runs, each instrumented with a different set of hardware counters.<n>Our approach matches bursts across executions by analyzing MPI structure, timing, and communication patterns.<n>The output is a new synthetic trace with all merged counters, which can be used both for HPC performance prediction and for conventional performance analysis.
  arXiv  Detail & Related papers  (2026-05-15T10:36:15Z)
• cuNNQS-SCI: A Fully GPU-Accelerated Framework for High-Performance Configuration Interaction Selection with Neural Network Quantum States [12.436495225801863]
  cuNNQS-SCI is a fully GPU-accelerated SCI framework designed to overcome these bottlenecks.<n>On an NVIDIA A100 cluster with 64 GPUs, cuNNQS-SCI achieves up to 2.32X end-to-end speedup.<n>It demonstrates excellent distributed performance, maintaining over 90% parallel efficiency in strong scaling tests.
  arXiv  Detail & Related papers  (2026-04-17T07:15:18Z)
• High Performance Quantum Emulation for Chemistry Applications with Hyperion [0.0]
  Hyperion is a massively parallel, GPU-accelerated quantum emulator architected to bypass the classical memory walls inherent in strongly correlated quantum chemistry simulations.<n> Hyperion offers a high-fidelity platform dedicated to the development of new quantum algorithms for chemistry.
  arXiv  Detail & Related papers  (2026-04-01T17:27:56Z)
• A Scalable Distributed Quantum Optimization Framework via Factor Graph Paradigm [46.08923284345648]
  We introduce a structure-aware framework for distributed quantum optimization.<n>We find that for a search space of size $N$, our framework achieves $O(sqrtN)$ query complexity up to processors and separator dependent factors.<n>We show that structure-aware decomposition delivers a practical path to scalable distributed quantum optimization on quantum networks.
  arXiv  Detail & Related papers  (2026-03-08T15:15:52Z)
• On-Chip Hardware-Aware Quantization for Mixed Precision Neural Networks [52.97107229149988]
  We propose an On-Chip Hardware-Aware Quantization framework, performing hardware-aware mixed-precision quantization on deployed edge devices. For efficiency metrics, we built an On-Chip Quantization Aware pipeline, which allows the quantization process to perceive the actual hardware efficiency of the quantization operator. For accuracy metrics, we propose Mask-Guided Quantization Estimation technology to effectively estimate the accuracy impact of operators in the on-chip scenario.
  arXiv  Detail & Related papers  (2023-09-05T04:39:34Z)
• Harnessing Deep Learning and HPC Kernels via High-Level Loop and Tensor Abstractions on CPU Architectures [67.47328776279204]
  This work introduces a framework to develop efficient, portable Deep Learning and High Performance Computing kernels. We decompose the kernel development in two steps: 1) Expressing the computational core using Processing Primitives (TPPs) and 2) Expressing the logical loops around TPPs in a high-level, declarative fashion. We demonstrate the efficacy of our approach using standalone kernels and end-to-end workloads that outperform state-of-the-art implementations on diverse CPU platforms.
  arXiv  Detail & Related papers  (2023-04-25T05:04:44Z)
• DeepGEMM: Accelerated Ultra Low-Precision Inference on CPU Architectures using Lookup Tables [49.965024476651706]
  DeepGEMM is a lookup table based approach for the execution of ultra low-precision convolutional neural networks on SIMD hardware. Our implementation outperforms corresponding 8-bit integer kernels by up to 1.74x on x86 platforms.
  arXiv  Detail & Related papers  (2023-04-18T15:13:10Z)
• MAPLE-X: Latency Prediction with Explicit Microprocessor Prior Knowledge [87.41163540910854]
  Deep neural network (DNN) latency characterization is a time-consuming process. We propose MAPLE-X which extends MAPLE by incorporating explicit prior knowledge of hardware devices and DNN architecture latency.
  arXiv  Detail & Related papers  (2022-05-25T11:08:20Z)
• Arline Benchmarks: Automated Benchmarking Platform for Quantum Compilers [0.0]
  Open-source software package, Arline Benchmarks, is designed to perform automated benchmarking of quantum compilers. We compare several quantum compilation frameworks based on a set of important metrics. We propose a concept of composite compilation pipeline that combines compiler-specific circuit optimizations in a single compilation stack.
  arXiv  Detail & Related papers  (2022-02-28T18:48:01Z)
• An Adaptive Device-Edge Co-Inference Framework Based on Soft Actor-Critic [72.35307086274912]
  High-dimension parameter model and large-scale mathematical calculation restrict execution efficiency, especially for Internet of Things (IoT) devices. We propose a new Deep Reinforcement Learning (DRL)-Soft Actor Critic for discrete (SAC-d), which generates the emphexit point, emphexit point, and emphcompressing bits by soft policy iterations. Based on the latency and accuracy aware reward design, such an computation can well adapt to the complex environment like dynamic wireless channel and arbitrary processing, and is capable of supporting the 5G URL
  arXiv  Detail & Related papers  (2022-01-09T09:31:50Z)
• Extending C++ for Heterogeneous Quantum-Classical Computing [56.782064931823015]
  qcor is a language extension to C++ and compiler implementation that enables heterogeneous quantum-classical programming, compilation, and execution in a single-source context. Our work provides a first-of-its-kind C++ compiler enabling high-level quantum kernel (function) expression in a quantum-language manner.
  arXiv  Detail & Related papers  (2020-10-08T12:49:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.