Browsing by Author "Nayak, Guruprasad"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Classifying multivariate time series by learning sequence-level discriminative patterns
    (2018-01-23) Nayak, Guruprasad; Mithal, Varun; Jia, Xiaowei; Kumar, Vipin
    Time series classification algorithms designed to use local context do not work on landcover classification problems where the instances of the two classes may often exhibit similar feature values due to the large natural variations in other land covers across the year and unrelated phenomena that they undergo. In this paper, we propose to learn discriminative patterns from the entire length of the time series, and use them as predictive features to identify the class of interest. We propose a novel neural network algorithm to learn the key signature of the class of interest as a function of the feature values together with the discriminative pattern made from that signature through the entire time series in a joint framework. We demonstrate the utility of this technique on the landcover classification application of burned area mapping that is of considerable societal importance.
  • Thumbnail Image
    Item
    Learning with Weak Supervision for Land Cover Mapping Problems
    (2020-01) Nayak, Guruprasad
    Land cover mapping is the task of generating maps of land use globally across time. The recent decades have seen an increasing availability of public satellite data sets with observations of the Earth at regular intervals of space and time. This coupled with the advances in machine learning and high performance computing provide an opportunity to automate the land cover mapping problem at scale. However, the availability of labeled data to train predictive models in this application is very limited, especially in the developing regions of the world, where accurate land cover maps are necessary for effective management of natural resources to sustain the rapid population growth in these regions. The need for labeled samples is further increased by: (1) Heterogeneity of land cover classes across space and time; (2) Increasing complexity of state-of-the-art predictive models and (3) Lack of sufficient samples at the required spatial and temporal resolutions. Since paucity of labeled data is a major problem in this domain, traditional machine learning algorithms that only rely on exact labeled data (strong supervision) have limited performance. This thesis investigates the use of weak supervision to mitigate the problem of not having sufficient samples with exact labels. In a weakly-supervised learning scenario, you have very few training samples that have exact labels corresponding to the target variable. However, you have plenty of weakly-labeled instances i.e you have an imperfect version of the target variable for these instances. The idea is that, by modeling the imperfection in the weak labels, we can mitigate the lack of (strongly-labeled) training data. We study three commonly-occurring sources of weak supervision for the land cover mapping problem: (1) Ordinal labels as weak supervision for regression (WORD); (2) Group-level labels as weak supervision for binary classification (WeaSL); and (3) Group-level labels with group-level features as weak supervision for binary classification (MultiRes). In each of these cases, we show that modeling the inexact nature of the weak supervision enables us to mitigate the lack of strong supervision. By extensive experiments on multiple data sets, we show that use of weak supervision (1) increases the generalizability of models trained with only strong supervision and (2) enables the use of more complex predictive models. In addition, since weak supervision is available in plenty, they provide a better representation of the class imbalance, when present in the population. WORD and WeaSL demonstrably optimize the performance of the model for rarity using weak supervision. Finally, although the data sets used in this thesis mainly come from the land cover problems of burned area mapping and urban mapping, the methods developed in this thesis are applicable to other domains as well, where similar forms of weak labels are available as demonstrated by experiments on data sets from other domains like natural language processing.
  • Thumbnail Image
    Item
    Mapping Burned Areas in Tropical forests using MODIS data
    (2016-09-02) Mithal, Varun; Nayak, Guruprasad; Khandelwal, Ankush; Kumar, Vipin; Nemani, Ramakrishna; Oza, Nikunj C.
    This paper presents a new burned area product for the tropical forests in South America and South-east Asia. The product is derived from Moderate Resolution Imaging Spectroradiometer (MODIS) multispectral surface reflectance data and Active Fire hotspots using a novel rare class detection framework that builds data-adaptive classification models for different spatial regions and land cover classes. Burned areas are reported for 9 MODIS tiles at a spatial resolution of 500 m in the study period from 2001 to 2014. The total burned area detected in the tropical forests of South America and South-east Asia during these years is 2,286,385 MODIS pixels (approximately 571 K sq. km.), which is more than three times compared to the estimates by the state-of-the art MODIS MCD64A1 (742,886 MODIS pixels). We also present validation of this burned area product using (i) manual inspection of Landsat false color composites before and after burn date, (ii) manual inspection of synchronized changes in vegetation index time series around the burn date, and (iii) comprehensive quantitative validation using MODIS-derived differenced Normalized Burn Ratio (dNBR). Our validation results indicate that the events reported in our product are indeed true burn events that are missed by the state-of-art burned area products.
  • Thumbnail Image
    Item
    Multiple Instance Learning for bags with Ordered instances
    (2017-06-07) Nayak, Guruprasad; Mithal, Varun; Kumar, Vipin
    Multiple Instance Learning (MIL) algorithms are designed for problems where labels are available for groups of instances, commonly referred to as bags. In this paper, we consider a new MIL prob- lem setting where instances in a bag are not ex- changeable, and a bijection exists between every pair of bags. We propose a neural network based MIL algorithm (MILOrd) that leverages the exis- tence of such a bijection when learning to discrim- inate bags. MILOrd has an input node for each in- stance in the bag, an output node that captures the bag level prediction, and a hidden layer that cap- tures the output from an instance level classifier for each instance in the bag. The bag level prediction is obtained by combining these hidden layer val- ues using a function that models the importance of each instance, unlike the traditional schemes where each instance is considered equal. We demonstrate the utility of the proposed algorithm on the prob- lem of burned area mapping using yearly bags com- posed of multispectral reflectance data for different time steps in the year. Our experiments show that MILOrd outperforms traditional MIL schemes that don’t account for the presence of a bijection.