arXiv:2608. 10738v1 Announce Type: new Abstract: We study the approximation of dynamical systems by semi-autonomous neural ordinary differential equations (SA-NODEs) over long time horizons.
arXiv:2511.08097v2 Announce Type: replace-cross Abstract: We consider a general infinite horizon Heterogeneous Restless multi-armed Bandit (RMAB). Heterogeneity is a fundamental problem for many real...
arXiv:2609.07247v1 Announce Type: new Abstract: Scaling the interaction horizon-the maximum number of environment interactions per episode-improves LLM agents on long-horizon tasks, and curriculum-ba...
arXiv:2607. 26370v1 Announce Type: cross Abstract: We propose a self-adaptive online learning for control method for tracking unknown target dynamics.
arXiv:2606. 09929v1 Announce Type: cross Abstract: Physical reservoir computing harnesses nonlinear mechanical dynamics but, by convention, freezes the substrate and trains only a linear readout, presuming the substrate is not usefully trainable.
arXiv:2604. 26256v2 Announce Type: replace Abstract: Reinforcement learning (RL) has become a critical paradigm for LLM post-training, yet the rollout phase -- accounting for 50--80% of total step time -- is bottlenecked by skewed generation: long-tailed trajectories indispensable for model performance block the entire training pipeline.
The paper introduces a conflict‑predictive variable horizon for distributed model predictive control in multi‑drone collision avoidance. Each drone locally estimates future conflicts using observed positions and confidence funnels, then selects the minimal horizon that covers the farthest predicted conflict, shrinking in clear air and expanding only when necessary. The authors prove that this adaptive horizon preserves recursive feasibility and asymptotic stability for linear models, and demonstrate in simulation that it reduces per‑step solver cost and total computation while maintaining separation on dense benchmarks.
arXiv:2606. 29519v1 Announce Type: new Abstract: Long-range learning is hard for recurrent networks trained with stochastic gradient descent, because the influence of a past input fades with the lag $\ell$, and if it fades too fast the dependence cannot be learned from finite data.
arXiv:2609.39888v2 Announce Type: new Abstract: Neural trajectory predictors can reach low prediction error while violating dynamics, actuator limits, or state constraints, especially when controls a...
arXiv:2607. 10362v1 Announce Type: new Abstract: Latent world models are trained to predict future states in a learned representation and are then deployed inside a planner that selects actions by simulating them forward.
arXiv:2609.13845v1 Announce Type: cross Abstract: World models trained with joint-embedding predictive architectures learn compact, structured latent representations from physical interaction, yet pl...
Long-range learning is hard for recurrent networks trained with stochastic gradient descent, because the influence of a past input fades with the lag $\ell$, and if it fades too fast the dependence cannot be learned from finite data. This fade is captured by an envelope $f(\ell)$.
The paper introduces OrbitTrace, a benchmark of 50 physics‑grounded compute‑availability traces from satellite orbits, and investigates whether specialized interruption‑resilient optimizers are needed when training is interrupted by predictable compute gaps. Experiments on CIFAR‑10/ResNet‑18 and GPT‑2/AdamW show that a strong checkpoint‑and‑resume baseline that preserves full optimizer state and indexes learning‑rate schedules in effective time matches uninterrupted training, rendering most availability‑aware methods unnecessary. Only in a narrow regime—large models with non‑persistable optimizer state and frequent short pauses—does reactive adaptation recover a modest portion of the state‑loss penalty, and even this benefit disappears for eclipse‑scale gaps.