Position: The Need for Ultrafast Training
This addresses the problem of limited adaptability in high-frequency systems for researchers and engineers in fields like quantum computing and fusion control, proposing a foundational but not yet implemented paradigm shift.
The paper argues for a shift from static inference-only accelerators to ultrafast on-chip learning in FPGAs to enable real-time adaptation in non-stationary, high-frequency environments, with applications in quantum error correction, plasma control, and autonomous experiments.
Domain-specialized FPGAs have delivered unprecedented performance for low-latency inference across scientific and industrial workloads, yet nearly all existing accelerators assume static models trained offline, relegating learning and adaptation to slower CPUs or GPUs. This separation fundamentally limits systems that must operate in non-stationary, high-frequency environments, where model updates must occur at the timescale of the underlying physics. In this paper, I argue for a shift from inference-only accelerators to ultrafast on-chip learning, in which both inference and training execute directly within the FPGA fabric under deterministic, sub-microsecond latency constraints. Bringing learning into the same real-time datapath as inference would enable closed-loop systems that adapt as fast as the physical processes they control, with applications spanning quantum error correction, cryogenic qubit calibration, plasma and fusion control, accelerator tuning, and autonomous scientific experiments. Enabling such regimes requires rethinking algorithms, architectures, and toolflows jointly, but promises to transform FPGAs from static inference engines into real-time learning machines.