arXiv subjects

Jan Finkbeiner

Publications and source records attributed to Jan Finkbeiner.

13 recordsLinked to original sources

arXiv2026

GRINQH: Graded Input-based Quantization Hierarchy for Efficient LLM Generation

Autoregressive decoding with LLMs is primarily bottlenecked by GPU memory bandwidth, especially in edge-computing settings. While quantization is essential for mitigating this bottleneck, most existing methods treat inference as a uniform process and fail to account for the asymmetry between the compute-bound prefill stage and the memory-bound decoding stage. We propose GRINQH (GRaded INput-based Quantization Hierarchy), a weight-only post-training quantization framework that accelerates decoding by unifying quantization and sparsification. GRINQH leverages activation magnitudes as a proxy for computational importance to dynamically assign weight channels to different precision levels, enabling flexible average bit widths during decoding. Evaluated on Llama3 and Qwen3 models, GRINQH outperforms state-of-the-art fixed- and mixed-precision baselines at comparable 3- and 4-bit settings, even enabling effective 2-bit generation. We experimentally verify theoretical speedups by leveraging a hierarchical nested memory layout for multi-precision storage in a custom GPU kernel. Ultimately, GRINQH establishes a new state-of-the-art Pareto frontier for LLM generation, enabling a dynamic trade-off between generation quality and inference speed.

cs.LG↗

arXiv2026

Learning to Remember, Learn, and Forget in Attention-Based Models

In-Context Learning (ICL) in transformers acts as an online associative memory and is believed to underpin their high performance on complex sequence processing tasks. However, in gated linear attention models, this memory has a fixed capacity and is prone to interference, especially for long sequences. We propose Palimpsa, a self-attention model that views ICL as a continual learning problem that must address a stability-plasticity dilemma. Palimpsa uses Bayesian metaplasticity, where the plasticity of each attention state is tied to an importance state grounded by a prior distribution that captures accumulated knowledge. We demonstrate that various gated linear attention models emerge as specific architecture choices and posterior approximations, and that Mamba2 is a special case of Palimpsa where forgetting dominates. This theoretical link enables the transformation of any non-metaplastic model into a metaplastic one, significantly expanding its memory capacity. Our experiments show that Palimpsa consistently outperforms baselines on the Multi-Query Associative Recall (MQAR) benchmark and on Commonsense Reasoning tasks.

cs.LG↗

arXiv2024

Optimal Gradient Checkpointing for Sparse and Recurrent Architectures using Off-Chip Memory

Recurrent neural networks (RNNs) are valued for their computational efficiency and reduced memory requirements on tasks involving long sequence lengths but require high memory-processor bandwidth to train. Checkpointing techniques can reduce the memory requirements by only storing a subset of intermediate states, the checkpoints, but are still rarely used due to the computational overhead of the additional recomputation phase. This work addresses these challenges by introducing memory-efficient gradient checkpointing strategies tailored for the general class of sparse RNNs and Spiking Neural Networks (SNNs). SNNs are energy efficient alternatives to RNNs thanks to their local, event-driven operation and potential neuromorphic implementation. We use the Intelligence Processing Unit (IPU) as an exemplary platform for architectures with distributed local memory. We exploit its suitability for sparse and irregular workloads to scale SNN training on long sequence lengths. We find that Double Checkpointing emerges as the most effective method, optimizing the use of local memory resources while minimizing recomputation overhead. This approach reduces dependency on slower large-scale memory access, enabling training on sequences over 10 times longer or 4 times larger networks than previously feasible, with only marginal time overhead. The presented techniques demonstrate significant potential to enhance scalability and efficiency in training sparse and recurrent networks across diverse hardware platforms, and highlights the benefits of sparse activations for scalable recurrent neural network training.

cs.NE↗

arXiv2024

On-Chip Learning via Transformer In-Context Learning

Autoregressive decoder-only transformers have become key components for scalable sequence processing and generation models. However, the transformer's self-attention mechanism requires transferring prior token projections from the main memory at each time step (token), thus severely limiting their performance on conventional processors. Self-attention can be viewed as a dynamic feed-forward layer, whose matrix is input sequence-dependent similarly to the result of local synaptic plasticity. Using this insight, we present a neuromorphic decoder-only transformer model that utilizes an on-chip plasticity processor to compute self-attention. Interestingly, the training of transformers enables them to ``learn'' the input context during inference. We demonstrate this in-context learning ability of transformers on the Loihi 2 processor by solving a few-shot classification problem. With this we emphasize the importance of pretrained models especially their ability to find simple, local, backpropagation free, learning rules enabling on-chip learning and adaptation in a hardware friendly manner.

cs.NE↗

arXiv2024

Analog In-Memory Computing Attention Mechanism for Fast and Energy-Efficient Large Language Models

Transformer networks, driven by self-attention, are central to Large Language Models. In generative Transformers, self-attention uses cache memory to store token projections, avoiding recomputation at each time step. However, GPU-stored projections must be loaded into SRAM for each new generation step, causing latency and energy bottlenecks. We present a custom self-attention in-memory computing architecture based on emerging charge-based memories called gain cells, which can be efficiently written to store new tokens during sequence generation and enable parallel analog dot-product computation required for self-attention. However, the analog gain cell circuits introduce non-idealities and constraints preventing the direct mapping of pre-trained models. To circumvent this problem, we design an initialization algorithm achieving text processing performance comparable to GPT-2 without training from scratch. Our architecture respectively reduces attention latency and energy consumption by up to two and five orders of magnitude compared to GPUs, marking a significant step toward ultra-fast, low-power generative Transformers.

cs.NE↗

arXiv2024

SNNAX -- Spiking Neural Networks in JAX

Spiking Neural Networks (SNNs) simulators are essential tools to prototype biologically inspired models and neuromorphic hardware architectures and predict their performance. For such a tool, ease of use and flexibility are critical, but so is simulation speed especially given the complexity inherent to simulating SNN. Here, we present SNNAX, a JAX-based framework for simulating and training such models with PyTorch-like intuitiveness and JAX-like execution speed. SNNAX models are easily extended and customized to fit the desired model specifications and target neuromorphic hardware. Additionally, SNNAX offers key features for optimizing the training and deployment of SNNs such as flexible automatic differentiation and just-in-time compilation. We evaluate and compare SNNAX to other commonly used machine learning (ML) frameworks used for programming SNNs. We provide key performance metrics, best practices, documented examples for simulating SNNs in SNNAX, and implement several benchmarks used in the literature.

cs.NE↗

arXiv2024

Efficient Event-Based Object Detection: A Hybrid Neural Network with Spatial and Temporal Attention

Event cameras offer high temporal resolution and dynamic range with minimal motion blur, making them promising for robust object detection. While Spiking Neural Networks (SNNs) on neuromorphic hardware are often considered for energy-efficient and low latency event-based data processing, they often fall short of Artificial Neural Networks (ANNs) in accuracy and flexibility. Here, we introduce Attention-based Hybrid SNN-ANN backbones for event-based object detection to leverage the strengths of both SNN and ANN architectures. A novel Attention-based SNN-ANN bridge module captures sparse spatial and temporal relations from the SNN layer and converts them into dense feature maps for the ANN part of the backbone. Additionally, we present a variant that integrates DWConvL-STMs to the ANN blocks to capture slower dynamics. This multi-timescale network combines fast SNN processing for short timesteps with long-term dense RNN processing, effectively capturing both fast and slow dynamics. Experimental results demonstrate that our proposed method surpasses SNN-based approaches by significant margins, with results comparable to existing ANN and RNN-based methods. Unlike ANN-only networks, the hybrid setup allows us to implement the SNN blocks on digital neuromorphic hardware to investigate the feasibility of our approach. Extensive ablation studies and implementation on neuromorphic hardware confirm the effectiveness of our proposed modules and architectural choices. Our hybrid SNN-ANN architectures pave the way for ANN-like performance at a drastically reduced parameter, latency, and power budget.

cs.CV↗

arXiv2024

The Ouroboros of Memristors: Neural Networks Facilitating Memristor Programming

Memristive devices hold promise to improve the scale and efficiency of machine learning and neuromorphic hardware, thanks to their compact size, low power consumption, and the ability to perform matrix multiplications in constant time. However, on-chip training with memristor arrays still faces challenges, including device-to-device and cycle-to-cycle variations, switching non-linearity, and especially SET and RESET asymmetry. To combat device non-linearity and asymmetry, we propose to program memristors by harnessing neural networks that map desired conductance updates to the required pulse times. With our method, approximately 95% of devices can be programmed within a relative percentage difference of +-50% from the target conductance after just one attempt. Our approach substantially reduces memristor programming delays compared to traditional write-and-verify methods, presenting an advantageous solution for on-chip training scenarios. Furthermore, our proposed neural network can be accelerated by memristor arrays upon deployment, providing assistance while reducing hardware overhead compared with previous works. This work contributes significantly to the practical application of memristors, particularly in reducing delays in memristor programming. It also envisions the future development of memristor-based machine learning accelerators.

cs.ET↗

arXiv2023

Harnessing Manycore Processors with Distributed Memory for Accelerated Training of Sparse and Recurrent Models

Current AI training infrastructure is dominated by single instruction multiple data (SIMD) and systolic array architectures, such as Graphics Processing Units (GPUs) and Tensor Processing Units (TPUs), that excel at accelerating parallel workloads and dense vector matrix multiplications. Potentially more efficient neural network models utilizing sparsity and recurrence cannot leverage the full power of SIMD processor and are thus at a severe disadvantage compared to today's prominent parallel architectures like Transformers and CNNs, thereby hindering the path towards more sustainable AI. To overcome this limitation, we explore sparse and recurrent model training on a massively parallel multiple instruction multiple data (MIMD) architecture with distributed local memory. We implement a training routine based on backpropagation through time (BPTT) for the brain-inspired class of Spiking Neural Networks (SNNs) that feature binary sparse activations. We observe a massive advantage in using sparse activation tensors with a MIMD processor, the Intelligence Processing Unit (IPU) compared to GPUs. On training workloads, our results demonstrate 5-10x throughput gains compared to A100 GPUs and up to 38x gains for higher levels of activation sparsity, without a significant slowdown in training convergence or reduction in final model performance. Furthermore, our results show highly promising trends for both single and multi IPU configurations as we scale up to larger model sizes. Our work paves the way towards more efficient, non-standard models via AI training hardware beyond GPUs, and competitive large scale SNN models.

cs.NE↗

arXiv2023

Generating Minimal Training Sets for Machine Learned Potentials

This letter presents a novel approach for identifying uncorrelated atomic configurations from extensive data sets with a non-standard neural network workflow known as random network distillation (RND) for training machine-learned inter-atomic potentials (MLPs). This method is coupled with a DFT workflow wherein initial data is generated with cheaper classical methods before only the minimal subset is passed to a more computationally expensive ab initio calculation. This benefits training not only by reducing the number of expensive DFT calculations required but also by providing a pathway to the use of more accurate quantum mechanical calculations for training. The method's efficacy is demonstrated by constructing machine-learned inter-atomic potentials for the molten salts KCl and NaCl. Our RND method allows accurate models to be fit on minimal data sets, as small as 32 configurations, reducing the required structures by at least one order of magnitude compared to alternative methods.

physics.comp-ph↗

arXiv2023

Online Transformers with Spiking Neurons for Fast Prosthetic Hand Control

Transformers are state-of-the-art networks for most sequence processing tasks. However, the self-attention mechanism often used in Transformers requires large time windows for each computation step and thus makes them less suitable for online signal processing compared to Recurrent Neural Networks (RNNs). In this paper, instead of the self-attention mechanism, we use a sliding window attention mechanism. We show that this mechanism is more efficient for continuous signals with finite-range dependencies between input and target, and that we can use it to process sequences element-by-element, this making it compatible with online processing. We test our model on a finger position regression dataset (NinaproDB8) with Surface Electromyographic (sEMG) signals measured on the forearm skin to estimate muscle activities. Our approach sets the new state-of-the-art in terms of accuracy on this dataset while requiring only very short time windows of 3.5 ms at each inference step. Moreover, we increase the sparsity of the network using Leaky-Integrate and Fire (LIF) units, a bio-inspired neuron model that activates sparsely in time solely when crossing a threshold. We thus reduce the number of synaptic operations up to a factor of $\times5.3$ without loss of accuracy. Our results hold great promises for accurate and fast online processing of sEMG signals for smooth prosthetic hand control and is a step towards Transformers and Spiking Neural Networks (SNNs) co-integration for energy efficient temporal signal processing.

cs.NE↗

arXiv2021

Efficient Data Selection Methods for the Development of Machine Learned Potentials

We present an investigation into data selection methods for the efficient sampling of configuration space as applied to the development of inter-atomic potentials for scale bridging in molecular dynamics (MD) simulations. This investigation suggests that the most efficient sampling techniques are those that incorporate information on an atomic level such as forces or atomic energies. Finally, we generate an inter-atomic potential for the a sodium chloride system using each data selection technique and find that the global selection methods result in non-physical simulations.

physics.comp-ph↗

arXiv2020

Single-Shot 3D Detection of Vehicles from Monocular RGB Images via Geometry Constrained Keypoints in Real-Time

In this paper we propose a novel 3D single-shot object detection method for detecting vehicles in monocular RGB images. Our approach lifts 2D detections to 3D space by predicting additional regression and classification parameters and hence keeping the runtime close to pure 2D object detection. The additional parameters are transformed to 3D bounding box keypoints within the network under geometric constraints. Our proposed method features a full 3D description including all three angles of rotation without supervision by any labeled ground truth data for the object's orientation, as it focuses on certain keypoints within the image plane. While our approach can be combined with any modern object detection framework with only little computational overhead, we exemplify the extension of SSD for the prediction of 3D bounding boxes. We test our approach on different datasets for autonomous driving and evaluate it using the challenging KITTI 3D Object Detection as well as the novel nuScenes Object Detection benchmarks. While we achieve competitive results on both benchmarks we outperform current state-of-the-art methods in terms of speed with more than 20 FPS for all tested datasets and image resolutions.

cs.CV↗