arXiv subjects

Xenofon Foukas

Publications and source records attributed to Xenofon Foukas.

6 recordsLinked to original sources

arXiv2026

Kairos: A Scalable Serving System for Physical AI

Physical AI is experiencing rapid growth with frontier foundation models increasing its capabilities across general environments. Physical AI tasks are characterized by inference properties that are markedly different from digital AI. They consist of multiple rounds of inference and action execution, generating a chunk of actions in each inference round, and asynchronously interleaving inference and execution. This makes existing digital AI serving systems unsuited for physical AI; a shortcoming that is critical for enabling their wide adoption, considering their size and the scale of the robot fleets they have to serve. To fill this gap, we design Kairos, the first multi-robot serving system that makes the generate-execute loop a first-class citizen, with active involvement in the execution phase. Across a wide range of physical AI models and robots, Kairos reduces the average end-to-end task latency by 31.8--66.5% over state-of-the-art digital AI serving practices, with gains scaling with the robot fleet size.

cs.RO↗

arXiv2026

Offload or Overload: A Platform Measurement Study of Mobile Robotic Manipulation Workloads

Mobile robotic manipulation--the ability of robots to navigate spaces and interact with objects--is a core capability of physical AI. Foundation models have led to breakthroughs in their performance, but at a significant computational cost. We present the first measurement study of mobile robotic manipulation workloads across onboard, edge, and cloud GPU platforms. We find that the full workload stack is infeasible to run on smaller onboard GPUs, while larger onboard GPUs drain robot batteries several hours faster. Offloading alleviates these constraints but introduces its own challenges, as additional network latency degrades task accuracy, and the bandwidth requirement makes naive cloud offloading impractical. Finally, we quantify opportunities and pitfalls of sharing compute across robot fleets. We believe our measurement study will be crucial to designing inference systems for mobile robots.

cs.RO↗

arXiv2025

Hades: Hierarchical Adaptable Decoding for Efficient and Elastic vRAN

In cellular networks, virtualized Radio Access Networks (vRANs) enable replacing traditional specialized hardware at cell sites with software running on commodity servers distributed across edge and remote clouds. However, some vRAN functions (e.g., forward error correction (FEC) decoding) require excessive edge compute resources due to their intensive computational demands and inefficiencies caused by workload fluctuations. This high demand for computational power significantly drives up the costs associated with edge computing, posing a major challenge for deploying 5G/6G vRAN solutions. To address this challenge, we propose Hades, a hierarchical architecture for vRAN that enables the distribution of uplink FEC decoding processing across edge and remote clouds. Hades refactors the vRAN stack and introduces mechanisms that allow controlling and managing the workload over these hierarchical cloud resources. More specifically, Hades splits the traditional non-stop run-to-completion iterative FEC decoding process into latency-critical early decoding iterations, i.e., related to MAC processing and early pre-parsing for content identification, and completion decoding iterations, i.e., decoding tasks with larger decoding delay budgets for final data bits extraction. This partitioning provides Hades the flexibility to utilize the available midhaul (MH) network for offloading the latency tolerant part of decoding to remote cloud instances, while performing time-sensitive decoding at the edge cloud locations for low-delay processing. Hades controls decoding load distribution between the edge and remote clouds, based on the edge decoding capacity and the offload network bandwidth, thus improving the utilization of edge compute.

cs.NI↗

arXiv2024

Distributed AI Platform for the 6G RAN

Cellular Radio Access Networks (RANs) are rapidly evolving towards 6G, driven by the need to reduce costs and introduce new revenue streams for operators and enterprises. In this context, AI emerges as a key enabler in solving complex RAN problems spanning both the management and application domains. Unfortunately, and despite the undeniable promise of AI, several practical challenges still remain, hindering the widespread adoption of AI applications in the RAN space. In this work, we attempt to shed light to these challenges and argue that existing approaches in addressing them are inadequate for realizing the vision of a truly AI-native 6G network. We propose a distributed AI platform architecture, tailored to the needs of an AI-native RAN.

cs.NI↗

arXiv2024

AutoSpec: Automated Generation of Neural Network Specifications

The increasing adoption of neural networks in learning-augmented systems highlights the growing need for model safety and robustness, especially in safety-critical domains. While recent advances in neural network verification offer formal guarantees on worst-case behavior, existing approaches require users to manually define model specifications, an error-prone, incomplete, and time-consuming process. In this paper, we present AutoSpec, the first comprehensive framework for automatically generating and evaluating neural network specifications for learning-augmented systems. AutoSpec introduces a tree-based algorithm that adaptively partitions the input space to generate specification sets aligned with model behavior, as well as a statistical certification framework that provides rigorous accuracy guarantees for each specification. We also propose a principled evaluation framework that defines interpretable metrics for specification accuracy and coverage, establishing a benchmark for future research. Experiments across four diverse applications show that AutoSpec outperforms both manually defined specifications and existing baseline algorithms, improving the F1 score by up to 53% over human-defined specifications and 73% over the strongest baseline.

cs.LG↗

arXiv2019

Iris: Deep Reinforcement Learning Driven Shared Spectrum Access Architecture for Indoor Neutral-Host Small Cells

We consider indoor mobile access, a vital use case for current and future mobile networks. For this key use case, we outline a vision that combines a neutral-host based shared small-cell infrastructure with a common pool of spectrum for dynamic sharing as a way forward to proliferate indoor small-cell deployments and open up the mobile operator ecosystem. Towards this vision, we focus on the challenges pertaining to managing access to shared spectrum (e.g., 3.5GHz US CBRS spectrum). We propose Iris, a practical shared spectrum access architecture for indoor neutral-host small-cells. At the core of Iris is a deep reinforcement learning based dynamic pricing mechanism that efficiently mediates access to shared spectrum for diverse operators in a way that provides incentives for operators and the neutral-host alike. We then present the Iris system architecture that embeds this dynamic pricing mechanism alongside cloud-RAN and RAN slicing design principles in a practical neutral-host design tailored for the indoor small-cell environment. Using a prototype implementation of the Iris system, we present extensive experimental evaluation results that not only offer insight into the Iris dynamic pricing process and its superiority over alternative approaches but also demonstrate its deployment feasibility.

cs.NI↗