arXiv subjects

Vijay Janapa Reddi

Publications and source records attributed to Vijay Janapa Reddi.

At least 37 records · Page 2Linked to original sources

arXiv2026

GenAI for Systems: Recurring Challenges and Design Principles from Software to Silicon

Generative AI is reshaping how computing systems are designed, optimized, and built, yet research remains fragmented across software, architecture, and chip design communities. This paper takes a cross-stack perspective, examining how generative models are being applied from code generation and distributed runtimes through hardware design space exploration to RTL synthesis, physical layout, and verification. Rather than reviewing each layer in isolation, we analyze how the same structural difficulties and effective responses recur across the stack. Our central finding is one of convergence. Despite the diversity of domains and tools, the field keeps encountering five recurring challenges (the feedback loop crisis, the tacit knowledge problem, trust and validation, co-design across boundaries, and the shift from determinism to dynamism) and keeps arriving at five design principles that independently emerge as effective responses (embracing hybrid approaches, designing for continuous feedback, separating concerns by role, matching methods to problem structure, and building on decades of systems knowledge). We organize these into a challenge--principle map that serves as a diagnostic and design aid, showing which principles have proven effective for which challenges across layers. Through concrete cross-stack examples, we show how systems navigate this map as they mature, and argue that the field needs shared engineering methodology, including common vocabularies, cross-layer benchmarks, and systematic design practices, so that progress compounds across communities rather than being rediscovered in each one. Our analysis covers more than 275 papers spanning eleven application areas across three layers of the computing stack, and distills open research questions that become visible only from a cross-layer vantage point.

cs.SE↗

arXiv2026

TinyTorch: Building Machine Learning Systems from First Principles

Machine learning education faces a fundamental gap: students learn algorithms without understanding the systems that execute them. They study gradient descent without measuring memory, attention mechanisms without analyzing O(N^2) scaling, optimizer theory without knowing why Adam requires 3x the memory of SGD. This "algorithm-systems divide" produces practitioners who can train models but cannot debug memory failures, optimize inference latency, or reason about deployment trade-offs--the very skills industry demands as "ML systems engineering." We present TinyTorch, a 20-module curriculum that closes this gap through "implementation-based systems pedagogy": students construct PyTorch's core components (tensors, autograd, optimizers, CNNs, transformers) in pure Python, building a complete framework where every operation they invoke is code they wrote. The design employs three patterns: "progressive disclosure" of complexity, "systems-first integration" of profiling from the first module, and "build-to-validate milestones" recreating 67 years of ML breakthroughs--from Perceptron (1958) through Transformers (2017) to MLPerf-style benchmarking. Requiring only 4GB RAM and no GPU, TinyTorch demonstrates that deep ML systems understanding is achievable without specialized hardware. The curriculum is available open-source at mlsysbook.ai/tinytorch.

cs.LG↗

arXiv2025

AI Benchmark Democratization and Carpentry

Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.

cs.AI↗

arXiv2025

Modeling and Optimizing Performance Bottlenecks for Neuromorphic Accelerators

Neuromorphic accelerators offer promising platforms for machine learning (ML) inference by leveraging event-driven, spatially-expanded architectures that naturally exploit unstructured sparsity through co-located memory and compute. However, their unique architectural characteristics create performance dynamics that differ fundamentally from conventional accelerators. Existing workload optimization approaches for neuromorphic accelerators rely on aggregate network-wide sparsity and operation counting, but the extent to which these metrics actually improve deployed performance remains unknown. This paper presents the first comprehensive performance bound and bottleneck analysis of neuromorphic accelerators, revealing the shortcomings of the conventional metrics and offering an understanding of what facets matter for workload performance. We present both theoretical analytical modeling and extensive empirical characterization of three real neuromorphic accelerators: Brainchip AKD1000, Synsense Speck, and Intel Loihi 2. From these, we establish three distinct accelerator bottleneck states, memory-bound, compute-bound, and traffic-bound, and identify which workload configuration features are likely to exhibit these bottleneck states. We synthesize all of our insights into the floorline performance model, a visual model that identifies performance bounds and informs how to optimize a given workload, based on its position on the model. Finally, we present an optimization methodology that combines sparsity-aware training with floorline-informed partitioning. Our methodology achieves substantial performance improvements at iso-accuracy: up to 3.86x runtime improvement and 3.38x energy reduction compared to prior manually-tuned configurations.

cs.AR↗

arXiv2025

Datasheets for Machine Learning Sensors

Machine learning (ML) is becoming prevalent in embedded AI sensing systems. These "ML sensors" enable context-sensitive, real-time data collection and decision-making across diverse applications ranging from anomaly detection in industrial settings to wildlife tracking for conservation efforts. As such, there is a need to provide transparency in the operation of such ML-enabled sensing systems through comprehensive documentation. This is needed to enable their reproducibility, to address new compliance and auditing regimes mandated in regulation and industry-specific policy, and to verify and validate the responsible nature of their operation. To address this gap, we introduce the datasheet for ML sensors framework. We provide a comprehensive template, collaboratively developed in academia-industry partnerships, that captures the distinct attributes of ML sensors, including hardware specifications, ML model and dataset characteristics, end-to-end performance metrics, and environmental impacts. Our framework addresses the continuous streaming nature of sensor data, real-time processing requirements, and embeds benchmarking methodologies that reflect real-world deployment conditions, ensuring practical viability. Aligned with the FAIR principles (Findability, Accessibility, Interoperability, and Reusability), our approach enhances the transparency and reusability of ML sensor documentation across academic, industrial, and regulatory domains. To show the application of our approach, we present two datasheets: the first for an open-source ML sensor designed in-house and the second for a commercial ML sensor developed by industry collaborators, both performing computer vision-based person detection.

cs.LG↗

arXiv2025

Adaptive Surrogate Gradients for Sequential Reinforcement Learning in Spiking Neural Networks

Neuromorphic computing systems are set to revolutionize energy-constrained robotics by achieving orders-of-magnitude efficiency gains, while enabling native temporal processing. Spiking Neural Networks (SNNs) represent a promising algorithmic approach for these systems, yet their application to complex control tasks faces two critical challenges: (1) the non-differentiable nature of spiking neurons necessitates surrogate gradients with unclear optimization properties, and (2) the stateful dynamics of SNNs require training on sequences, which in reinforcement learning (RL) is hindered by limited sequence lengths during early training, preventing the network from bridging its warm-up period. We address these challenges by systematically analyzing surrogate gradient slope settings, showing that shallower slopes increase gradient magnitude in deeper layers but reduce alignment with true gradients. In supervised learning, we find no clear preference for fixed or scheduled slopes. The effect is much more pronounced in RL settings, where shallower slopes or scheduled slopes lead to a 2.1x improvement in both training and final deployed performance. Next, we propose a novel training approach that leverages a privileged guiding policy to bootstrap the learning process, while still exploiting online environment interactions with the spiking policy. Combining our method with an adaptive slope schedule for a real-world drone position control task, we achieve an average return of 400 points, substantially outperforming prior techniques, including Behavioral Cloning and TD3BC, which achieve at most --200 points under the same conditions. This work advances both the theoretical understanding of surrogate gradient learning in SNNs and practical training methodologies for neuromorphic controllers demonstrated in real-world robotic systems.

cs.AI↗

arXiv2025

SwizzlePerf: Hardware-Aware LLMs for GPU Kernel Performance Optimization

Large language models (LLMs) have shown progress in GPU kernel performance engineering using inefficient search-based methods that optimize around runtime. Any existing approach lacks a key characteristic that human performance engineers rely on for near-optimal utilization -- hardware-awareness. By leveraging the workload's specific memory access patterns, architecture specifications, filtered profiling logs, and reflections on historical performance, we can make software-level optimizations that are tailored to the underlying hardware. SwizzlePerf automatically generates spatial optimizations for GPU kernels on disaggregated architectures by giving LLMs explicit hardware-awareness. For a GEMM kernel, SwizzlePerf takes less than 5 minutes to generate the same hardware-specific optimal swizzling pattern that took expert performance engineers 2 weeks to find. On a suite of 10 diverse ML and Science kernels, SwizzlePerf can generate swizzling patterns for 9 of the kernels that achieve up to a 2.06x speedup and 70% improvement in L2 hit rate. This work is the first of many steps toward systematically creating hardware-aware LLM performance engineering agents.

cs.DC↗

arXiv2025

From Seed to Harvest: Augmenting Human Creativity with AI for Red-teaming Text-to-Image Models

Text-to-image (T2I) models have become prevalent across numerous applications, making their robust evaluation against adversarial attacks a critical priority. Continuous access to new and challenging adversarial prompts across diverse domains is essential for stress-testing these models for resilience against novel attacks from multiple vectors. Current techniques for generating such prompts are either entirely authored by humans or synthetically generated. On the one hand, datasets of human-crafted adversarial prompts are often too small in size and imbalanced in their cultural and contextual representation. On the other hand, datasets of synthetically-generated prompts achieve scale, but typically lack the realistic nuances and creative adversarial strategies found in human-crafted prompts. To combine the strengths of both human and machine approaches, we propose Seed2Harvest, a hybrid red-teaming method for guided expansion of culturally diverse, human-crafted adversarial prompt seeds. The resulting prompts preserve the characteristics and attack patterns of human prompts while maintaining comparable average attack success rates (0.31 NudeNet, 0.36 SD NSFW, 0.12 Q16). Our expanded dataset achieves substantially higher diversity with 535 unique geographic locations and a Shannon entropy of 7.48, compared to 58 locations and 5.28 entropy in the original dataset. Our work demonstrates the importance of human-machine collaboration in leveraging human creativity and machine computational capacity to achieve comprehensive, scalable red-teaming for continuous T2I model safety evaluation.

cs.LG↗

arXiv2025

Multi-Agent Reinforcement Learning for Sample-Efficient Deep Neural Network Mapping

Mapping deep neural networks (DNNs) to hardware is critical for optimizing latency, energy consumption, and resource utilization, making it a cornerstone of high-performance accelerator design. Due to the vast and complex mapping space, reinforcement learning (RL) has emerged as a promising approach-but its effectiveness is often limited by sample inefficiency. We present a decentralized multi-agent reinforcement learning (MARL) framework designed to overcome this challenge. By distributing the search across multiple agents, our framework accelerates exploration. To avoid inefficiencies from training multiple agents in parallel, we introduce an agent clustering algorithm that assigns similar mapping parameters to the same agents based on correlation analysis. This enables a decentralized, parallelized learning process that significantly improves sample efficiency. Experimental results show our MARL approach improves sample efficiency by 30-300x over standard single-agent RL, achieving up to 32.61x latency reduction and 16.45x energy-delay product (EDP) reduction under iso-sample conditions.

cs.LG↗

arXiv2025

COSMIC: Enabling Full-Stack Co-Design and Optimization of Distributed Machine Learning Systems

Large-scale machine learning models necessitate distributed systems, posing significant design challenges due to the large parameter space across distinct design stacks. Existing studies often focus on optimizing individual system aspects in isolation. This work challenges this limitation and introduces COSMIC, a full-stack distributed machine learning systems environment enabling end-to-end simulation and agent-based design space exploration. To facilitate efficient exploration and optimization across the entire stack, we introduce Parameter Set Architecture-an abstraction concept analogous to the instruction set architecture-abstracting away configuration complexities of agent-based search methods. Case studies demonstrate COSMIC's ability to consolidate parameters across multiple layers of design abstraction, discovering eight non-obvious high-performance system configurations across four transformer-based models with up to 175 billion parameters. By optimizing across the stack, COSMIC full-stack optimization delivers 1.50-48.41x higher performance compared to the isolated single-stack optimization.

cs.DC↗

arXiv2025

Generative AI in Embodied Systems: System-Level Analysis of Performance, Efficiency and Scalability

Embodied systems, where generative autonomous agents engage with the physical world through integrated perception, cognition, action, and advanced reasoning powered by large language models (LLMs), hold immense potential for addressing complex, long-horizon, multi-objective tasks in real-world environments. However, deploying these systems remains challenging due to prolonged runtime latency, limited scalability, and heightened sensitivity, leading to significant system inefficiencies. In this paper, we aim to understand the workload characteristics of embodied agent systems and explore optimization solutions. We systematically categorize these systems into four paradigms and conduct benchmarking studies to evaluate their task performance and system efficiency across various modules, agent scales, and embodied tasks. Our benchmarking studies uncover critical challenges, such as prolonged planning and communication latency, redundant agent interactions, complex low-level control mechanisms, memory inconsistencies, exploding prompt lengths, sensitivity to self-correction and execution, sharp declines in success rates, and reduced collaboration efficiency as agent numbers increase. Leveraging these profiling insights, we suggest system optimization strategies to improve the performance, efficiency, and scalability of embodied agents across different paradigms. This paper presents the first system-level analysis of embodied AI agents, and explores opportunities for advancing future embodied system design.

cs.RO↗

arXiv2025

Machine Learning Fleet Efficiency: Analyzing and Optimizing Large-Scale Google TPU Systems with ML Productivity Goodput

Recent years have seen the emergence of machine learning (ML) workloads deployed in warehouse-scale computing (WSC) settings, also known as ML fleets. As the computational demands placed on ML fleets have increased due to the rise of large models and growing demand for ML applications, it has become increasingly critical to measure and improve the efficiency of such systems. However, there is not yet an established methodology to characterize ML fleet performance and identify potential performance optimizations accordingly. This paper presents a large-scale analysis of an ML fleet based on Google's TPUs, introducing a framework to capture fleet-wide efficiency, systematically evaluate performance characteristics, and identify optimization strategies for the fleet. We begin by defining an ML fleet, outlining its components, and analyzing an example Google ML fleet in production comprising thousands of accelerators running diverse workloads. Our study reveals several critical insights: first, ML fleets extend beyond the hardware layer, with model, data, framework, compiler, and scheduling layers significantly impacting performance; second, the heterogeneous nature of ML fleets poses challenges in characterizing individual workload performance; and third, traditional utilization-based metrics prove insufficient for ML fleet characterization. To address these challenges, we present the "ML Productivity Goodput" (MPG) metric to measure ML fleet efficiency. We show how to leverage this metric to characterize the fleet across the ML system stack. We also present methods to identify and optimize performance bottlenecks using MPG, providing strategies for managing warehouse-scale ML systems in general. Lastly, we demonstrate quantitative evaluations from applying these methods to a real ML fleet for internal-facing Google TPU workloads, where we observed tangible improvements.

cs.LG↗

arXiv2025

Materiality and Risk in the Age of Pervasive AI Sensors

Artificial intelligence (AI) systems connected to sensor-laden devices are becoming pervasive, which has notable implications for a range of AI risks, including to privacy, the environment, autonomy and more. There is therefore a growing need for increased accountability around the responsible development and deployment of these technologies. Here we highlight the dimensions of risk associated with AI systems that arise from the material affordances of sensors and their underlying calculative models. We propose a sensor-sensitive framework for diagnosing these risks, complementing existing approaches such as the US National Institute of Standards and Technology AI Risk Management Framework and the European Union AI Act, and discuss its implementation. We conclude by advocating for increased attention to the materiality of algorithmic systems, and of on-device AI sensors in particular, and highlight the need for development of a sensor design paradigm that empowers users and communities and leads to a future of increased fairness, accountability and transparency.

cs.CY↗

arXiv2025

A2Perf: Real-World Autonomous Agents Benchmark

Autonomous agents and systems cover a number of application areas, from robotics and digital assistants to combinatorial optimization, all sharing common, unresolved research challenges. It is not sufficient for agents to merely solve a given task; they must generalize to out-of-distribution tasks, perform reliably, and use hardware resources efficiently during training and inference, among other requirements. Several methods, such as reinforcement learning and imitation learning, are commonly used to tackle these problems, each with different trade-offs. However, there is a lack of benchmarking suites that define the environments, datasets, and metrics which can be used to provide a meaningful way for the community to compare progress on applying these methods to real-world problems. We introduce A2Perf--a benchmark with three environments that closely resemble real-world domains: computer chip floorplanning, web navigation, and quadruped locomotion. A2Perf provides metrics that track task performance, generalization, system resource efficiency, and reliability, which are all critical to real-world applications. Using A2Perf, we demonstrate that web navigation agents can achieve latencies comparable to human reaction times on consumer hardware, reveal reliability trade-offs between algorithms for quadruped locomotion, and quantify the energy costs of different learning approaches for computer chip-design. In addition, we propose a data cost metric to account for the cost incurred acquiring offline data for imitation learning and hybrid algorithms, which allows us to better compare these approaches. A2Perf also contains several standard baselines, enabling apples-to-apples comparisons across methods and facilitating progress in real-world autonomy. As an open-source benchmark, A2Perf is designed to remain accessible, up-to-date, and useful to the research community over the long term.

cs.LG↗

arXiv2025

MLPerf Power: Benchmarking the Energy Efficiency of Machine Learning Systems from Microwatts to Megawatts for Sustainable AI

Rapid adoption of machine learning (ML) technologies has led to a surge in power consumption across diverse systems, from tiny IoT devices to massive datacenter clusters. Benchmarking the energy efficiency of these systems is crucial for optimization, but presents novel challenges due to the variety of hardware platforms, workload characteristics, and system-level interactions. This paper introduces MLPerf Power, a comprehensive benchmarking methodology with capabilities to evaluate the energy efficiency of ML systems at power levels ranging from microwatts to megawatts. Developed by a consortium of industry professionals from more than 20 organizations, MLPerf Power establishes rules and best practices to ensure comparability across diverse architectures. We use representative workloads from the MLPerf benchmark suite to collect 1,841 reproducible measurements from 60 systems across the entire range of ML deployment scales. Our analysis reveals trade-offs between performance, complexity, and energy efficiency across this wide range of systems, providing actionable insights for designing optimized ML solutions from the smallest edge devices to the largest cloud infrastructures. This work emphasizes the importance of energy efficiency as a key metric in the evaluation and comparison of the ML system, laying the foundation for future research in this critical area. We discuss the implications for developing sustainable AI solutions and standardizing energy efficiency benchmarking for ML systems.

cs.AR↗

arXiv2025

SocratiQ: A Generative AI-Powered Learning Companion for Personalized Education and Broader Accessibility

Traditional educational approaches often struggle to provide personalized and interactive learning experiences on a scale. In this paper, we present SocratiQ, an AI-powered educational assistant that addresses this challenge by implementing the Socratic method through adaptive learning technologies. The system employs a novel Generative AI-based learning framework that dynamically creates personalized learning pathways based on student responses and comprehension patterns. We provide an account of our integration methodology, system architecture, and evaluation framework, along with the technical and pedagogical challenges encountered during implementation and our solutions. Although our implementation focuses on machine learning systems education, the integration approaches we present can inform similar efforts across STEM fields. Through this work, our goal is to advance the understanding of how generative AI technologies can be designed and systematically incorporated into educational resources.

cs.CY↗

arXiv2025

NeuroBench: A Framework for Benchmarking Neuromorphic Computing Algorithms and Systems

Neuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks standardized benchmarks, making it difficult to accurately measure technological advancements, compare performance with conventional methods, and identify promising future research directions. Prior neuromorphic computing benchmark efforts have not seen widespread adoption due to a lack of inclusive, actionable, and iterative benchmark design and guidelines. To address these shortcomings, we present NeuroBench: a benchmark framework for neuromorphic computing algorithms and systems. NeuroBench is a collaboratively-designed effort from an open community of researchers across industry and academia, aiming to provide a representative structure for standardizing the evaluation of neuromorphic approaches. The NeuroBench framework introduces a common set of tools and systematic methodology for inclusive benchmark measurement, delivering an objective reference framework for quantifying neuromorphic approaches in both hardware-independent (algorithm track) and hardware-dependent (system track) settings. In this article, we outline tasks and guidelines for benchmarks across multiple application domains, and present initial performance baselines across neuromorphic and conventional approaches for both benchmark tracks. NeuroBench is intended to continually expand its benchmarks and features to foster and track the progress made by the research community.

cs.AI↗

arXiv2025

QuArch: A Question-Answering Dataset for AI Agents in Computer Architecture

We introduce QuArch, a dataset of 1500 human-validated question-answer pairs designed to evaluate and enhance language models' understanding of computer architecture. The dataset covers areas including processor design, memory systems, and performance optimization. Our analysis highlights a significant performance gap: the best closed-source model achieves 84% accuracy, while the top small open-source model reaches 72%. We observe notable struggles in memory systems, interconnection networks, and benchmarking. Fine-tuning with QuArch improves small model accuracy by up to 8%, establishing a foundation for advancing AI-driven computer architecture research. The dataset and leaderboard are at https://harvard-edge.github.io/QuArch/.

cs.AR↗