Recent advancements in deep convolutional neural networks have significantly improved the performance of saliency prediction. However, the manual configuration of the neural network architectures requires domain knowledge expertise and can still be time-consuming and error-prone. To solve this, we propose a new Neural Architecture Search (NAS) framework for saliency prediction with two contributions. Firstly, a supernet for saliency prediction is built with a weight-sharing network containing all candidate architectures, by integrating a dynamic convolution into the encoder-decoder in the supernet, termed SalNAS. Secondly, despite the fact that SalNAS is highly efficient (20.98 million parameters), it can suffer from the lack of generalization. To solve this, we propose a self-knowledge distillation approach, termed Self-KD, that trains the student SalNAS with the weighted average information between the ground truth and the prediction from the teacher model. The teacher model, while sharing the same architecture, contains the best-performing weights chosen by cross-validation. Self-KD can generalize well without the need to compute the gradient in the teacher model, enabling an efficient training system. By utilizing Self-KD, SalNAS outperforms other state-of-the-art saliency prediction models in most evaluation rubrics across seven benchmark datasets while being a lightweight model. The code will be available at https://github.com/chakkritte/SalNAS

3D Euclidean symmetry equivariant neural networks have demonstrated notable success in modeling complex physical systems. We introduce a framework for relaxed $E(3)$ graph equivariant neural networks that can learn and represent symmetry breaking within continuous groups. Building on the existing e3nn framework, we propose the use of relaxed weights to allow for controlled symmetry breaking. We show empirically that these relaxed weights learn the correct amount of symmetry breaking.

This document presents the Center for Robot Assisted Search And Rescue - Uncrewed Aerial Systems - Disaster Response Overhead Inspection Dataset (CRASAR-U-DROIDs) for building damage assessment and spatial alignment collected from small uncrewed aerial systems (sUAS) geospatial imagery. This dataset is motivated by the increasing use of sUAS in disaster response and the lack of previous work in utilizing high-resolution geospatial sUAS imagery for machine learning and computer vision models, the lack of alignment with operational use cases, and with hopes of enabling further investigations between sUAS and satellite imagery. The CRASAR-U-DRIODs dataset consists of fifty-two (52) orthomosaics from ten (10) federally declared disasters (Hurricane Ian, Hurricane Ida, Hurricane Harvey, Hurricane Idalia, Hurricane Laura, Hurricane Michael, Musset Bayou Fire, Mayfield Tornado, Kilauea Eruption, and Champlain Towers Collapse) spanning 67.98 square kilometers (26.245 square miles), containing 21,716 building polygons and damage labels, and 7,880 adjustment annotations. The imagery was tiled and presented in conjunction with overlaid building polygons to a pool of 130 annotators who provided human judgments of damage according to the Joint Damage Scale. These annotations were then reviewed via a two-stage review process in which building polygon damage labels were first reviewed individually and then again by committee. Additionally, the building polygons have been aligned spatially to precisely overlap with the imagery to enable more performant machine learning models to be trained. It appears that CRASAR-U-DRIODs is the largest labeled dataset of sUAS orthomosaic imagery.

LiDAR-based SLAM is a core technology for autonomous vehicles and robots. Despite the intense research activity in this field, each proposed system uses a particular sensor post-processing pipeline and a single map representation format. The present work aims at introducing a revolutionary point of view for 3D LiDAR SLAM and localization: (1) using view-based maps as the fundamental representation of maps ("simple-maps"), which can then be used to generate arbitrary metric maps optimized for particular tasks; and (2) by introducing a new framework in which mapping pipelines can be defined without coding, defining the connections of a network of reusable blocks much like deep-learning networks are designed by connecting layers of standardized elements. Moreover, the idea of including the current linear and angular velocity vectors as variables to be optimized within the ICP loop is also introduced, leading to superior robustness against aggressive motion profiles without an IMU. The presented open-source ecosystem, released to ROS 2, includes tools and prebuilt pipelines covering all the way from data acquisition to map editing and visualization, real-time localization, loop-closure detection, or map georeferencing from consumer-grade GNSS receivers. Extensive experimental validation reveals that the proposal compares well to, or improves, former state-of-the-art (SOTA) LiDAR odometry systems, while also successfully mapping some hard sequences where others diverge. A proposed self-adaptive configuration has been used, without parameter changes, for all 3D LiDAR datasets with sensors between 16 and 128 rings, extensively tested on 83 sequences over more than 250~km of automotive, hand-held, airborne, and quadruped LiDAR datasets, both indoors and outdoors. The open-sourced implementation is available online at https://github.com/MOLAorg/mola

What makes the PC the superior platform for personal computing? Sometimes that choice can be daunting though. Qualcomm's new Snapdragon X Elite just launched in Microsoft's debut Copilot_ laptops, coming out swinging against Intel's Core Ultra chip. With the launch of Ryzen AI 300-series today, you're now faced with a third choice. Fear not, as I've just finished testing performance testing of AMD's Ryzen AI 9 HX 370, its new flagship processor for Copilot laptops.

Optimization problems are pervasive in sectors from manufacturing and distribution to healthcare. However, most such problems are still solved heuristically by hand rather than optimally by state-of-the-art solvers because the expertise required to formulate and solve these problems limits the widespread adoption of optimization tools and techniques. We introduce a Large Language Model (LLM)-based system designed to formulate and solve (mixed integer) linear programming problems from their natural language descriptions. Our system is capable of developing mathematical models, writing and debugging solver code, evaluating the generated solutions, and improving efficiency and correctness of its model and code based on these evaluations.

In recent years, there is increased interest in foundation models for geoscience due to vast amount of earth observing satellite imagery. Existing remote sensing foundation models make use of the various sources of spectral imagery to create large models pretrained on masked reconstruction task. The embeddings from these foundation models are then used for various downstream remote sensing applications. In this paper we propose a foundational modeling framework for remote sensing geoscience applications, that goes beyond these traditional single modality masked autoencoder family of foundation models. This framework leverages the knowledge guided principles that the spectral imagery captures the impact of the physical drivers on the environmental system, and that the relationship between them is governed by the characteristics of the system. Specifically, our method, called MultiModal Variable Step Forecasting (MM-VSF), uses mutlimodal data (spectral imagery and weather) as its input and a variable step forecasting task as its pretraining objective. In our evaluation we show forecasting of satellite imagery using weather can be used as an effective pretraining task for foundation models. We further show the effectiveness of the embeddings from MM-VSF on the downstream task of pixel wise crop mapping, when compared with a model trained in the traditional setting of single modality input and masked reconstruction based pretraining.

Solar energy is one of the most promising renewable energy resources. Forecasting photovoltaic power generation is an important way to increase photovoltaic penetration. However, the task of photovoltaic forecasting is complicated due to its property of uncertainty, especially in specific regions during the foggy winter. This paper proposes a novel model to accomplish the problem. A developed entropy is created to qualify the uncertainty during the foggy winter. The clustering method and modified retention network are applied to reduce complexity and forecast, respectively. We adopt an optimization to optimize the hyperparameters. Results are validated from the multivariate forecasting model using the dataset from a photovoltaic power station in Jiangsu Province, China. Experiments show that the proposed model improves the forecasting accuracy compared to various models during the foggy winter.

This research presents an innovative use of parallel computing with the ARIMA (AutoRegressive Integrated Moving Average) model to forecast energy consumption in Peru's Puno region. The study conducts a thorough and multifaceted analysis, focusing on the execution speed, prediction accuracy, and scalability of both sequential and parallel implementations. A significant emphasis is placed on efficiently managing large datasets. The findings demonstrate notable improvements in computational efficiency and data processing capabilities through the parallel approach, all while maintaining the accuracy and integrity of predictions. This new method provides a versatile and reliable solution for real-time predictive analysis and enhances energy resource management, which is particularly crucial for developing areas. In addition to highlighting the technical advantages of parallel computing in this field, the study explores its practical impacts on energy planning and sustainable development in regions like Puno.

Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, 306 N Wright St, Urbana, IL 61801, USA (Dated: July 30, 2024) A central question in deep learning is how deep neural networks (DNNs) learn features. This collective effect of non-linearity, noise, learning rate, width, depth, and numerous other parameters, has eluded first-principles theories which are built from microscopic neuronal dynamics. Here we present a noise-non-linearity phase diagram that highlights where shallow or deep layers learn features more effectively. We then propose a macroscopic mechanical theory of feature learning that accurately reproduces this phase diagram, offering a clear intuition for why and how some DNNs are "lazy" and some are "active", and relating the distribution of feature learning over layers with test accuracy. Deep neural networks (DNNs) progressively compute propose a macroscopic theory of feature learning in deep, features from which the final layer generates predictions.