VLVladimir Loncar

Vladimir Loncar

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Vladimir Loncar has a total of 112 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    SparsePixels: efficient convolution for sparse data on FPGAs

    by Ho Fung Tsoi, Dylan Rankin, Vladimir Loncar, Philip Harris

    Inference of standard convolutional neural networks (CNNs) on FPGAs often incurs high latency and a long initiation interval due to the deep nested loops required to densely convolve every input pixel regardless of its feature value. However, input features can be spatially sparse in some image data, where semantic information may occupy only a small fraction of the pixels and most computation would be wasted on empty regions. In this work, we introduce SparsePixels, a framework that implements sparse convolution on FPGAs by selectively retaining and computing on a small subset of active input pixels while ignoring the rest, reducing the per-layer compute from O ( H W K 2 C in C out ) to O ( n 2 C in C out ) and the storage from O ( H W C ) to O ( n C ) when n 2 ≪ H W . Because computation always runs over a single pre-specified pixel budget that is frozen at synthesis, the inference latency is data-independent and constant at runtime. We show that, for identifying neutrino interactions in naturally sparse LArTPC images with 4k pixels, a standard CNN with a compact size of 4k parameters incurs an inference latency of 48.665 μ s on an FPGA, whereas a sparse CNN of the same base architecture, computing on less than 1% of the input pixels, achieves a × 73 speedup to 0.665 μ s with resource utilization well within on-chip budgets, trading only a small percent-level performance loss. This work aims to benefit future algorithm development for efficient data readout in modern experiments with strict latency requirements of microseconds or below.

  • Open Access

    PQuantML: a tool for end-to-end hardware-aware model compression

    by Roope Niemi, Anastasiia Petrovych, Arghya Ranjan Das, Enrico Lupi, Chang Sun, Dimitrios Danopoulos, Marlon Joshua Helbing, Mia Liu, Sebastian Dittmeier, Michael Kagan, et al.

    PQuantML is a new open-source, hardware-aware neural network model compression library tailored to end-to-end workflows. Motivated by the need to deploy performant models to environments with strict latency constraints, PQuantML simplifies training of compressed models by providing a unified interface to apply pruning and quantization, either jointly or individually. The library implements multiple pruning methods with different granularities, as well as fixed-point quantization with support for high-granularity quantization (HGQ). We evaluate PQuantML on the jet substructure classification task, so-called jet tagging, an on-edge problem related to real-time large Hadron collider data processing. Using various pruning methods with fixed-point quantization, PQuantML achieves substantial parameter and bit-width reductions while maintaining accuracy. We benchmark PQuantML’s pruning and fixed-point quantization against QKeras, and validate that its integrated HGQ algorithm implementation reproduces the standalone HGQ library results.

  • Open Access

    Enabling low-latency machine learning on radiation-hard FPGAs with hls4ml

    by Katya Govorkova, Julián García Pardiñas, Vladimir Loncar, Sebastian Schmitt, Victoria Nguyen, Loris Martinazzoli, Eluned A Smith, Marco Pizzichemi

    This paper presents an end-to-end demonstration of a viable, ultra-fast, radiation-hard machine learning (ML) application on FPGAs, which could be used in future high-energy physics experiments. We present a three-fold contribution, with the PicoCal calorimeter, planned for the LHCb Upgrade II experiment, used as a test case. First, we develop a lightweight autoencoder to compress a 32-sample timing readout, representative of that of the PicoCal, into a two-dimensional latent space. Second, we introduce a systematic, hardware-aware quantization strategy and show that the model can be reduced to 10-bit weights with minimal performance loss. Third, as a barrier to the adoption of on-detector ML is the lack of support for radiation-hard FPGAs in the High-Energy Physics community’s standard ML synthesis tool, hls4ml, we develop a new backend for this library. This new back-end enables the automatic translation of ML models into High-Level Synthesis (HLS) projects for the Microchip PolarFire family of FPGAs, one of the few commercially available and radiation hard FPGAs. We present the synthesis of the autoencoder on a target PolarFire FPGA, which indicates that a latency of 25 ns can be achieved. We show that the resources utilized are low enough that the model can be placed within the inherently protected logic of the FPGA. Our extension to hls4ml is a significant contribution, paving the way for broader adoption of ML on FPGAs in high-radiation environments.


  • Open Access

    wa-hls4ml: A Benchmark and Surrogate Models for hls4ml Resource and Latency Estimation

    by Benjamin Hawks, Jason Weitz, Dmitri Demler, Karla Tame-Narvaez, Dennis Plotnikov, Mohammad Mehdi Rahimifar, Hamza Ezzaoui Rahali, Audrey C. Therrien, Donovan Sproule, Elham E. Khoda, et al.

    As machine learning (ML) is increasingly implemented in hardware to address real-time challenges in scientific applications, the development of advanced toolchains has significantly reduced the time required to iterate on various designs. These advancements have solved major obstacles, but also exposed new challenges. For example, processes that were not previously considered bottlenecks, such as hardware synthesis, are becoming limiting factors in the rapid iteration of designs. To mitigate these emerging constraints, multiple efforts have been undertaken to develop an ML-based surrogate model that estimates resource usage of ML accelerator architectures. We introduce wa-hls4ml, a benchmark for ML accelerator resource and latency estimation, and its corresponding initial dataset of over 680,000 fully connected and convolutional neural networks, all synthesized using hls4ml and targeting Xilinx FPGAs. The benchmark evaluates the performance of resource and latency predictors against several common ML model architectures, primarily originating from scientific domains, as exemplar models, and the average performance across a subset of the dataset. Additionally, we introduce GNN- and transformer-based surrogate models that predict latency and resources for ML accelerators. We present the architecture and performance of the models and find that the models generally predict latency and resources for the 75th percentile within several percent of the synthesized resources on the synthetic test dataset.


  • Top Cited Papers
    Open Access

    hls4ml: A Flexible, Open Source Platform for Deep Learning Acceleration on Reconfigurable Hardware

    by Jan-Frederik Schulte, Benjamin Ramhorst, Chang Sun, Jovan Mitrevski, Nicolò Ghielmetti, Enrico Lupi, Dimitrios Danopoulos, Vladimir Lončar, Javier Duarte, David Burnette, et al.

    We present hls4ml , a free and open source platform that translates machine learning (ML) models from modern deep learning frameworks into high-level synthesis (HLS) code that can be integrated into full designs for field-programmable gate arrays (FPGAs) or application-specific integrated circuits (ASICs). With its flexible and modular design, hls4ml supports a large number of deep learning frameworks and can target HLS compilers from several vendors, including Vitis HLS, Intel oneAPI and Catapult HLS. Together with a wider eco-system for software-hardware co-design, hls4ml has enabled the acceleration of ML inference in a wide range of commercial and scientific applications where low latency, resource usage, and power consumption are critical. In this article, we describe the structure and functionality of the hls4ml platform. The overarching design considerations for the generated HLS code are discussed, together with selected performance results.


  • Open Access

    It’s not a FAD: first demonstration of flows for unsupervised anomaly detection at 40 MHz for use at the Large Hadron Collider

    by Francesco Vaselli, Chang Sun, Dimitrios Danopoulos, Roope Oskari Niemi, Maciej Mikolaj Glowacki, Katya Govorkova, Felice Pantaleo, Thea Klaeboe Aarrestad, Vladimir Loncar, Maurizio Pierini

    We present the first implementation of a Continuous Normalizing Flow (CNF) model for unsupervised anomaly detection within the realistic, high-rate environment of the Large Hadron Collider's L1 trigger systems. While CNFs typically define an anomaly score via a probabilistic likelihood, calculating this score requires solving an Ordinary Differential Equation, a procedure too complex for FPGA deployment. To overcome this, we propose a novel, hardware-friendly anomaly score defined as the squared norm of the model's vector field output. This score is based on the intuition that anomalous events require a larger transformation by the flow, and it is shown to be physically interpretable as the norm of the input features for our specific training choice. Our model, trained via Flow Matching on Standard Model data, is synthesized for an FPGA using the hls4ml and da4ml libraries. We demonstrate that our approach effectively identifies a variety of beyond-the-Standard-Model signatures with performance comparable to existing machine learning-based triggers. The algorithm achieves a latency of a few hundred nanoseconds, or even less when using advanced quantization techniques, and requires minimal FPGA resources, establishing CNFs as a viable new tool for real-time, data-driven discovery at 40 MHz.

  • Top Cited Papers
    Open Access

    da4ml: Distributed Arithmetic for Real-time Neural Networks on FPGAs

    by Chang Sun, Zhiqiang Que, Vladimir Loncar, Wayne Luk, Maria Spiropulu

    Neural networks with a latency requirement on the order of microseconds, like the ones used at the CERN Large Hadron Collider, are typically deployed on FPGAs fully unrolled and pipelined. A bottleneck for the deployment of such neural networks is area utilization, which is directly related to the required constant matrix-vector multiplication (CMVM) operations. In this work, we propose an efficient algorithm for implementing CMVM operations with distributed arithmetic on FPGAs that simultaneously optimizes for area consumption and latency. The algorithm achieves resource reduction similar to state-of-the-art algorithms while being significantly faster to compute. The proposed algorithm is open sourced and integrated into the hls4ml library, a free and open source library for running real-time neural network inference on FPGAs. We show that the proposed algorithm can reduce on-chip resources by up to a third for realistic, highly quantized neural networks while simultaneously reducing latency, enabling the implementation of previously infeasible networks.


  • Open Access

    HGQ: High Granularity Quantization for Real-time Neural Networks on FPGAs

    by Chang Sun, Zhiqiang Que, Thea Aarrestad, Vladimir Loncar, Jennifer Ngadiuba, Wayne Luk, Maria Spiropulu


  • Open Access

    Enabling Low-Latency Machine learning on Radiation-Hard FPGAs with hls4ml

    by Katya Govorkova, Julian Garcia Pardinas, Vladimir Loncar, Victoria Nguyen, Sebastian Schmitt, Marco Pizzichemi, Loris Martinazzoli, Eluned Anne Smith

    This paper presents the first demonstration of a viable, ultra-fast, radiation-hard machine learning (ML) application on FPGAs, which could be used in future high-energy physics experiments. We present a three-fold contribution, with the PicoCal calorimeter, planned for the LHCb Upgrade II experiment, used as a test case. First, we develop a lightweight autoencoder to compress a 32-sample timing readout, representative of that of the PicoCal, into a two-dimensional latent space. Second, we introduce a systematic, hardware-aware quantization strategy and show that the model can be reduced to 10-bit weights with minimal performance loss. Third, as a barrier to the adoption of on-detector ML is the lack of support for radiation-hard FPGAs in the High-Energy Physics community's standard ML synthesis tool, hls4ml, we develop a new backend for this library. This new back-end enables the automatic translation of ML models into High-Level Synthesis (HLS) projects for the Microchip PolarFire family of FPGAs, one of the few commercially available and radiation hard FPGAs. We present the synthesis of the autoencoder on a target PolarFire FPGA, which indicates that a latency of 25 ns can be achieved. We show that the resources utilized are low enough that the model can be placed within the inherently protected logic of the FPGA. Our extension to hls4ml is a significant contribution, paving the way for broader adoption of ML on FPGAs in high-radiation environments.

  • Open Access

    wa-hls4ml and lui-gnn: A Benchmark and GNN based Surrogate Model for hls4ml Resource and Latency Estimation

    by Benjamin Hawks, Dennis Plotnikov, Nhan Tran, Karla Tame-Narvaez, Mohammad Mehdi Rahimifar, Hamza Ezzaoui Rahali, Audrey C. Therrien, Giuseppe Di Guglielmo, Javier Duarte, Vladimir Loncar

    As machine learning (ML) increasingly serves as a tool for addressing real-time challenges in scientific applications, the development of advanced tooling has significantly reduced the time required to iterate on various designs. These advancements have solved major obstacles, but also exposed new challenges. For example, processes that were not previously considered bottlenecks, such as model synthesis, are now becoming limiting factors in the rapid iteration of designs. To reduce these emerging constraints, multiple efforts are being launched toward designing an ML-based surrogate model that estimates resource usage of synthesized accelerator architectures. This model would reduce the design iteration time, especially when designing within a set of given hardware constraints. This approach shows considerable potential, but as it stands, the effort is early and would benefit from coordination and standardization to assist future work as it emerges. We introduce wa-hls4ml, a benchmark for ML accelerator resource and latency estimation, and its corresponding initial dataset of more than 100,000 fully connected neural networks, all synthesized using hls4ml and targeting Xilinx FPGAs. The benchmark evaluates the performance of resource and latency predictors against several common ML model architectures, primarily originating from scientific domains, as exemplar models, as well as the average performance across a subset of the dataset. We measure the performance of a given predictor model through multiple metrics, including R2 score and SMAPE on regression tasks, as well as inference time to further characterize the estimator under test. Additionally, we introduce the latency/utilization inference graph neural network (lui-gnn), a surrogate model that uses a graph neural network to represent input architectures in the form of a directed graph. We present the architecture and performance of the model and find that lui-gnn generally predicts latency and utilization for the 75% quantile within several percent of the synthesized resources on the synthetic test dataset, indicating that this approach of estimating resource and latency via a surrogate models has promise and warrants further research.