Scheduling

We study offline scheduling for large language model (LLM) serving under a fixed KV-cache memory budget, where requests have heterogeneous prompt (prefill) and response (decode) lengths. Prompt tokens determine initial KV-cache usage, while each generated token further increases memory consumption, creating dynamic memory constraints during autoregressive decoding. Given a backlog of n requests arriving together, the goal is to form mixed prefill and decode batches over time to minimize total end-to-end latency. We show that heterogeneous prompt lengths fundamentally change the scheduling problem: the problem is NP-hard, and standard policies such as first-come-first-served, shortest-output-first, and total-size-based prioritization can have unbounded approximation ratios. We propose Sorted-F, a scheduling algorithm that repeatedly forms feasible batches using an F-metric that balances batch size against downstream decode cost. We prove that Sorted-F achieves a constant-factor approximation guarantee in the offline/backlogged model. We also develop practical implementations, including an exact dynamic program for small instances and scalable local-search and greedy heuristics for larger instances, as well as LP-guided and receding-horizon variants. Experiments on public workloads that combine short conversations and long-document summarization show that F-metric-based scheduling consistently reduces latency relative to standard baselines and remains close to the LP relaxation lower bound for tractable instances.

In standard reinforcement learning (RL) settings, the interaction between the agent and the environment is typically modeled as a Markov decision process (MDP), which assumes that the agent observes the system state instantaneously, selects an action without delay, and executes it immediately. In real-world dynamic environments, such as cyber-physical systems, this assumption often breaks down due to delays in the interaction between the agent and the system. These delays can vary stochastically over time and are typically unobservable when deciding on an action. Existing methods deal with this uncertainty conservatively by assuming a known fixed upper bound on the delay, even if the delay is often much lower. In this work, we introduce the interaction layer, a general framework that enables agents to adaptively handle unobservable and time-varying delays. Specifically, the agent generates a matrix of possible future actions, anticipating a horizon of potential delays, to handle both unpredictable delays and lost action packets sent over networks. Building on this framework, we develop a model-based algorithm, Actor-Critic with Delay Adaptation (ACDA), which dynamically adjusts to delay patterns. Our method significantly outperforms state-of-the-art approaches across a wide range of locomotion benchmark environments, including real-world measured delays.

Real-time market prediction services need correct predictions before a decision deadline; a correct prediction delivered late is not usable. TIP-Search studies time-predictable inference scheduling over fixed market predictors under uncertain load. It filters conformal latency-quantile feasible models, dispatches over finite workers, and uses shielded constrained online experts to trade accuracy, queue pressure, and deadline risk. On the optimized deployable pool, TIP-Search reaches 0.994 raw accuracy and 0.991 timely accuracy. On official TLOB FI-2010 h=10, TIP-Search++ raises timely accuracy from 0.156 to 0.239 and deadline satisfaction from 0.391 to 0.962. In matched h10 profiled systems replay, OCO-ACPO reaches 0.303 timely accuracy and 0.951 deadline satisfaction, with paired gains over RAMSIS/SneakPeek/utility-style comparators of $+0.00285$ timely accuracy ($p=0.0118$) and $+0.0146$ deadline satisfaction ($p=1.5{\times}10^{-5}$). SA-OCO-ACPO improves timely/deadline service by 0.188--0.417 over CPO under nonstationary stress. The claim is a systems scheduling result, not a broad LOB classifier leaderboard.

Neuromorphic computing aims to improve the efficiency of artificial neural networks by taking inspiration from biological neurons and leveraging temporal sparsity, spatial sparsity, and compute near/in memory. Although these approaches have shown efficiency gains, training these spiking neural networks (SNN) remains difficult. The original attempts at converting trained conventional analog neural networks (ANN) to SNNs used the rate of binary spikes to represent neuron activations. This required many simulation time steps per inference, which degraded efficiency. Intel's Loihi 2 is a neuromorphic platform that supports graded spikes which can be used to represent changes in neuron activation. In this work, we use Loihi 2's graded spikes to develop a method for converting ANN networks to spiking networks, which exploits temporal and spatial sparsity. We evaluated the performance of this network on Loihi 2 and compared it to NVIDIA's Jetson Xavier edge AI platform. The results show that neuromorphic approaches achieve significant improvements in efficiency and latency (energy-delay product) over existing solutions.

We study budget pacing in repeated first-price auctions when an advertiser's private-value distributions change over time and the stationary competing-bid distribution is unknown. We ask how a feasible expenditure plan should enter online bid shading, learning, and hard budget control. We establish a plan-to-performance decomposition for a plan-driven projected-dual policy. The policy uses any feasible expenditure plan as a soft target, learns an unknown stationary competing-bid CDF from thresholds revealed after each auction, and enforces the campaign budget on every sample path. Against a distribution-informed expected-budget fluid benchmark, the uniform-plan reward gap is $O(\sqrt T)+O(\mathcal W_T)$, where $\mathcal W_T$ measures heterogeneity in private-value distributions. With a supplied feasible plan, the global gap decomposes into a one-sided $O(\sqrt{T})$ fixed-plan execution term and a plan-mismatch term bounded by $(b/2a)PlanError$. The same analysis provides guarantees for strict and relaxed period-cap comparators, exact recovery of the global benchmark under a specific allowance vector, and separate lower bounds establishing the necessity of the temporal-heterogeneity and Plan Error terms. An upstream planner can translate forecasts or managerial priorities into a feasible spending trajectory, while the online controller adapts bids using realized thresholds and expenditures. The guarantee is modular: it evaluates the final normalized or projected plan through $PlanError$. A specific forecasting model can be linked to the guarantee by establishing how its primitive estimation errors propagate to this plan-quality metric.

Increasingly, recommender systems are tasked with improving users' long-term satisfaction. In this context, we study a content exploration task, which we formalize as a bandit problem with delayed rewards. There is an apparent trade-off in choosing the learning signal: waiting for the full reward to become available might take several weeks, slowing the rate of learning, whereas using short-term proxy rewards reflects the actual long-term goal only imperfectly. First, we develop a predictive model of delayed rewards that incorporates all information obtained to date. Rewards as well as shorter-term surrogate outcomes are combined through a Bayesian filter to obtain a probabilistic belief. Second, we devise a bandit algorithm that quickly learns to identify content aligned with long-term success using this new predictive model. We prove a regret bound for our algorithm that depends on the Value of Progressive Feedback, an information-theoretic metric that captures the quality of short-term leading indicators that are observed prior to the long-term reward. We apply our approach to a podcast recommendation problem, where we seek to recommend shows that users engage with repeatedly over two months. We empirically validate that our approach significantly outperforms methods that optimize for short-term proxies or rely solely on delayed rewards, as demonstrated by an A/B test in a recommendation system that serves hundreds of millions of users.

We consider the problem of scheduling in multi-class, parallel-server queuing systems with uncertain rewards from job-server assignments. In this scenario, jobs incur holding costs while awaiting completion, and job-server assignments yield observable stochastic rewards with unknown mean values. The mean rewards for job-server assignments are assumed to follow a bilinear model with respect to features that characterize jobs and servers. Our objective is to minimize regret by maximizing the cumulative reward of job-server assignments over a time horizon, while keeping the total job holding cost bounded to ensure the stability of the queueing system. This problem is motivated by applications requiring resource allocation in network systems. A central challenge is to control the tradeoff between reward maximization and fair allocation for the stability of the underlying queuing system (i.e., maximizing network throughput). To address this challenge, we propose a scheduling algorithm based on a weighted proportional fair criteria augmented with marginal costs for reward maximization, incorporating a bandit algorithm tailored for bilinear rewards. Our algorithm admits a regret--queue length tradeoff. For any fixed control parameter $V>0$, it ensures a uniform expected queue length and time-average holding-cost bounds. For a target horizon $T$, choosing $V_T=Θ(\sqrt{IT})$ at initialization yields $\widetilde O((\sqrt I+d^2)\sqrt T+1/δ)$ regret. Under this regret-optimized tuning, the corresponding expected queue length and time-average holding-cost bounds remain uniform over the execution time and scales as $O(\sqrt{IT}+1/δ)$ and $O(\sqrt{IT}/δ)$, respectively.

Combinatorial scheduling poses a significant challenge for language models, requiring them to identify feasible solutions within exponentially large search spaces while satisfying complex constraints. This challenge is especially pronounced in resource-constrained settings, where larger language models are impractical and selection is limited to smaller models which often fail to preserve feasibility when scheduling directly from natural language. To address these limitations, we introduce SDDL, a neuro-symbolic framework that translates natural-language scheduling problems into compact, solver-aligned representations of tasks, resources, constraints, and objectives, while delegating low-level modeling and search to a deterministic compiler and external solver. On a 300-instance, multi-family subset of scheduling problems, SDDL improves independently verified feasibility for every resource-constrained model tested. The two strongest SDDL configurations reach 55.3% and 28.3%, up from direct-generation baselines of 23.7% and 1.3% and solver-code baselines of 21.7% and 7.0%, with a 0.0% median optimality gap among feasible schedules. By expressing problem structure rather than generating solutions or solver code, SDDL enables smaller models to approach the strongest evaluated direct- and solver-code configurations, including substantially larger frontier models.