arXiv:2601.04270v1 Announce Type: new Abstract: Deep learning optimization exhibits structure that is not captured by worst-case gradient bounds. Empirically, gradients along training trajectories are often temporally predictable and evolve within a low-dimensional subspace. In this work we formalize this observation through a measurable framework for predictable gradient manifolds. We introduce two computable quantities: a prediction-based path length that measures how well gradients can be forecast from past information, and a predictable rank that quantifies the intrinsic temporal dimension […]

Ver mais

This paper focuses on the scalable robot learning for manipulation in the dexterous robot arm-hand systems, where the remote human-robot interactions via augmented reality (AR) are established to collect the expert demonstration data for improving efficiency. In such a system, we present a unified framework to address the general manipulation task problem. Specifically, the proposed method consists of two phases: i) In the first phase for pretraining, the policy is created in a behavior cloning (BC) manner, through […]

Ver mais

Concurrency is the act of having your computer do multiple things at the same time. If you’ve heard a lot of talk about asyncio being added to Python but are curious how it compares to other concurrency methods or are wondering what concurrency is and how it might speed up your program, you’ve come to the right place. In this course, you’ll learn the following: How I/O bound programs are effected by latency Which concurrent programming patterns to […]

Ver mais

In this month’s episode of Functional Futures, our guest is Erik Svedäng, a game designer who has created many board and video games. Among them is Else Heart.Break(), a puzzle video game with its own programming language. He is also the creator of Carp, a statically-typed lisp for real-time applications. In the episode, we talk about game design, game development, and how Carp enables developers to build performant games while keeping true to functional programming idioms. As always, […]

Ver mais

Electricity use of AI coding agents Previous work estimating the energy and water cost of LLMs has generally focused on the cost per prompt using a consumer-level system such as ChatGPT. Simon P. Couch notes that coding agents such as Claude Code use way more tokens in response to tasks, often burning through many thousands of tokens of many tool calls. As a heavy Claude Code user, Simon estimates his own usage at the equivalent of 4,400 “typical […]

Ver mais

arXiv:2601.07957v1 Announce Type: new Abstract: Maize disease classification plays a vital role in mitigating yield losses and ensuring food security. However, the deployment of traditional disease detection models in resource-constrained environments, such as those using smartphones and drones, faces challenges due to high computational costs. To address these challenges, we propose LWMSCNN-SE, a lightweight convolutional neural network (CNN) that integrates multi-scale feature extraction, depthwise separable convolutions, and squeeze-and-Excitation (SE) attention mechanisms. This novel combination enables the model to […]

Ver mais

High-quality kernel is critical for scalable AI systems, and enabling LLMs to generate such code would advance AI development. However, training LLMs for this task requires sufficient data, a robust environment, and the process is often vulnerable to reward hacking and lazy optimization. In these cases, models may hack training rewards and prioritize trivial correctness over meaningful speedup. In this paper, we systematically study reinforcement learning (RL) for kernel generation. We first design KernelGYM, a robust distributed GPU […]

Ver mais

Spatio-Temporal (ST) Foundation Models (STFMs) promise cross-dataset generalization, yet joint ST pretraining is computationally expensive and grapples with the heterogeneity of domain-specific spatial patterns. Substantially extending our preliminary conference version, we present FactoST-v2, an enhanced factorized framework redesigned for full weight transfer and arbitrary-length generalization. FactoST-v2 decouples universal temporal learning from domain-specific spatial adaptation. The first stage pretrains a minimalist encoder-only backbone using randomized sequence masking to capture invariant temporal dynamics, enabling probabilistic quantile prediction across variable horizons. […]

Ver mais

In most enterprise Databricks environments (like in MSC or large analytics ecosystems), system tables such as system.jobrunlogs or system.cluster_events may be restricted or disabled due to security or governance policies. However, tracking cluster utilization and cost is crucial for : Understanding how efficiently jobs use compute Identifying idle clusters or cost leaks Forecasting infrastructure budget Building custom cost dashboards This blog demonstrates a step-by-step approach to compute cluster utilization and cost using only Databricks REST APIs — no […]

Ver mais