Sharing knowledge for multiple related machine learning tasks is an effective strategy to improve the generalization performance. In this paper, we investigate knowledge sharing across categories for action recognition in videos. The motivation is that many action categories are related, where common motion pattern are shared among them (e.g. diving and high jump share the jump motion). We propose a new multi-task learning method to learn latent tasks shared across categories, and reconstruct a classifier for each category from these latent tasks. Compared to previous methods, our approach has two advantages: (1) The learned latent tasks correspond to basic motion patterns instead of full actions, thus enhancing discrimination power of the classifiers. (2) Categories are selected to share information with a sparsity regularizer, avoiding falsely forcing all categories to share knowledge. Experimental results on multiple public data sets show that the proposed approach can effectively transfer knowledge between different action categories to improve the performance of conventional single task learning methods.
Remotely monitoring changes in central U.S. grasslands is challenging because these landscapes tend to respond quickly to disturbances and changes in weather. Such dynamic responses influence nutrient cycling, greenhouse gas contributions, habitat availability for wildlife, and other ecosystem processes and services. Traditionally, coarse-resolution satellite data acquired at daily intervals have been used for monitoring. Recently, the harmonized Landsat-8 and Sentinel-2 (HLS) data increased the temporal frequency of the data. Here we investigated if the increased data frequency provided adequate observations to characterize highly dynamic grassland processes. We evaluated HLS data available for 2016 to (1) determine if data from Sentinel-2 contributed to an improvement in characterizing landscape processes over Landsat-8 data alone, and (2) quantify how observation frequency impacted results. Specifically, we investigated into estimating annual vegetation phenology, detecting burn scars from fire, and modeling within-season wetland hydroperiod and growth of aquatic vegetation. We observed increased sensitivity to the start of the growing season (SOST) with the HLS data. Our estimates of the grassland SOST compared well with ground estimates collected at a phenological camera site. We used the Continuous Change Detection and Classification (CCDC) algorithm to assess if the HLS data improved our detection of burn scars following grassland fires and found that detection was considerably influenced by the seasonal timing of the fires. The grassland burned in early spring recovered too quickly to be detected as change events by CCDC; instead, the spectral characteristics following these fires were incorporated as part of the ongoing time-series models. In contrast, the spectral effects from late-season fires were detected both by Landsat-8 data and HLS data. For wetland-rich areas, we used a modified version of the CCDC algorithm to track within-season dynamics of water and aquatic vegetation. The addition of Sentinel-2 data provided the potential to build full time series models to better distinguish different wetland types, suggesting that the temporal density of data was sufficient for within-season characterization of wetland dynamics. Although the different data frequency, in both the spatial and temporal dimensions, could cause inconsistent model estimation or sensitivity sometimes; overall, the temporal frequency of the HLS data improved our ability to track within-season grassland dynamics and improved results for areas prone to cloud contamination. The results suggest a greater frequency of observations, such as from harmonizing data across all comparable Landsat and Sentinel sensors, is still needed. For our study areas, at least a 3-day revisit interval during the early growing season (weeks 14–17) is required to provide a >50% probability of obtaining weekly clear observations.
Multi-task learning (MTL) methods have shown promising performance by learning multiple relevant tasks simultaneously, which exploits to share useful information across relevant tasks. Among various MTL methods, clustered multi-task learning (CMTL) assumes that all tasks can be clustered into groups and attempts to learn the underlying cluster structure from the training data. In this paper, we present a new approach for CMTL, called flexible clustered multi-task (FCMTL), in which the cluster structure is learned by identifying representative tasks. The new approach allows an arbitrary task to be described by multiple representative tasks, effectively soft-assigning a task to multiple clusters with different weights. Unlike existing counterpart, the proposed approach is more flexible in that (a) it does not require clusters to be disjoint, (b) tasks within one particular cluster do not have to share information to the same extent, and (c) the number of clusters is automatically inferred from data. Computationally, the proposed approach is formulated as a row-sparsity pursuit problem. We validate the proposed FCMTL on both synthetic and real-world data sets, and empirical results demonstrate that it outperforms many existing MTL methods.
The environmental degradation is the deterioration of the environment through depletion of resources which includes all the biotic and abiotic element that form our surrounding that is air, water, soil, pant animals, and all other living and non-living element of the planet of earth. The major factor of environmental degradation is human (modern urbanization, industrialization, overpopulation growth, deforestation, etc.) and natural (flood, typhoons, droughts, rising temperatures, fires, etc.) cause. Today, different kinds of human activities are the main reasons for environmental degradation. The automobile and industries increase the number of poisonous gases like SOx, NOx, CO, and smoke in the atmosphere. Therefore, the government must enhance filling the gap in the legal system to avoid illegal activities. This chapter discusses the impact of environmental degradation with its future impacts, city planners, industry, and resource managers plans to be considered to mitigate the long term effects of developmental environmental degradation.
The objective of this work was to elucidate NO 3 -supply, Cl -toxicity, and Cl -/NO 3 -interaction in Glycine max and Glycine soja under salt stress. G. max cultivars (Lee68 and Jackson) and G. soja accessions (BB52 and N23227) with different salt tolerance were chosen as the experimental materials. Effects of low (0.75 mmol/L), normal (7.5 mmol/L), and high (15
To explore the influences of different cultivated areas on the chemical profiles of Eucommia ulmoides leaves (EUL) and rapidly authenticate its geographical origins, 187 samples from 13 provinces in China were systematically investigated using three data fusion strategies (low, mid, and high level) combined with two discrimination model algorithms (partial least squares discrimination analysis; random forest, RF). RF models constructed by high-level data fusion with different modes of different spectral data (Fourier transform near-infrared spectrum and attenuated total reflection Fourier transform mid-infrared spectrum) were most suitable for identifying EULs from different geographical origins. The accuracy rates of calibration and validation set were 92.86% and 93.44%, respectively. In addition, climate parameters were systematically investigated the cluster difference in our study. Some interesting and novel information could be found from the clustering tree diagram of hierarchical cluster analysis. The Xinjiang Autonomous Region (Region 5) located in the high latitude area was the only region in the middle temperate zone of all sample collection areas in which the samples belonged to an individual class no matter their distance in the tree diagram. The samples were from a relatively high elevation in the Shennongjia Forest District in Hubei Province (>1200 m), which is the main difference from the samples from Xiangyang City (78 m). Thus, the sample clusters from region 9 are different from the sample clusters from other regions. The results would provide a reference for further research to those samples from the special cluster.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.