An official website of the United States government
Graph-guided semi-supervised learning (SSL) has gained popularity in several network science applications, including biological, social, and financial ones. SSL becomes particularly challenging when the available nodal labels are scarce, what motivates naturally the active learning (AL) paradigm. AL seeks the most informative nodes to label in order to effectively estimate the nodal values of unobserved nodes. It is also referred to as active sampling, and boils down to learning the sought function mapping, and an acquisition function (AF) to identify the next node(s) to sample. To learn the mapping, this work leverages an adaptive Bayesian model comprising an ensemble (E) of Gaussian Processes (GPs) with enhanced expressiveness of the function space. Unlike most alternatives, the EGP model relies only on the one-hop connectivity of each node. Capitalizing on this EGP model, a suite of novel and intuitive AFs are developed to guide the active sampling process. These AFs are then combined with weights that are adapted incrementally to further robustify performance. Numerical tests on real and synthetic datasets corroborate the merits of the novel methods.
more »
« less
Online Graph-Guided Inference Using Ensemble Gaussian Processes of Egonet Features
Graph-guided semi-supervised learning (SSL) and inference has emerged as an attractive research field thanks to its documented impact in a gamut of application domains, including transportation and power networks, biological, social, environmental, and financial ones. Distinct from SSL approaches that yield point estimates of the variables to be inferred, the present work puts forth a Bayesian interval learning framework that utilizes Gaussian processes (GPs) to allow for uncertainty quantification – a key component in safety-critical applications. An ensemble (E) of GPs is employed to offer an expressive model of the learning function that is updated incrementally as nodal observations become available – what caters also for delay-sensitive settings. For the first time in graph-guided SSL and inference, egonet features per node are utilized as input to the EGP learning function to account for higher order interactions than the one-hop connectivity of each node. Further enhancing these attributes through random features that encrypt sensitive information per node offers scalability and privacy for the EGP-based learning approach. Numerical tests on real and synthetic datasets corroborate the effectiveness of the novel method.
more »
« less
- PAR ID:
- 10380650
- Date Published:
- Journal Name:
- Conference record Asilomar Conference on Signals Systems Computers
- ISSN:
- 1058-6393
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Bayesian optimization (BO) has well-documented merits for optimizing black-box functions with an expensive evaluation cost. Such functions emerge in applications as diverse as hyperparameter tuning, drug discovery, and robotics. BO hinges on a Bayesian surrogate model to sequentially select query points so as to balance exploration with exploitation of the search space. Most existing works rely on a single Gaussian process (GP) based surrogate model, where the kernel function form is typically preselected using domain knowledge. To bypass such a design process, this paper leverages an ensemble (E) of GPs to adaptively select the surrogate model fit on-the-fly, yielding a GP mixture posterior with enhanced expressiveness for the sought function. Acquisition of the next evaluation input using this EGP-based function posterior is then enabled by Thompson sampling (TS) that requires no additional design parameters. To endow function sampling with scalability, random feature-based kernel approximation is leveraged per GP model. The novel EGP-TS readily accommodates parallel operation. To further establish convergence of the proposed EGP-TS to the global optimum, analysis is conducted based on the notion of Bayesian regret for both sequential and parallel settings. Tests on synthetic functions and real-world applications showcase the merits of the proposed method.more » « less
-
Graph-guided learning has well-documented impact in a gamut of network science applications. A prototypical graph-guided learning task deals with semi-supervised learning over graphs, where the goal is to predict the nodal values or labels of unobserved nodes, by leveraging a few nodal observations along with the underlying graph structure. This is particularly challenging under privacy constraints or generally when acquiring nodal observations incurs high cost. In this context, the present work puts forth a Bayesian graph-driven self-supervised learning (Self-SL) approach that: (i) learns powerful nodal embeddings emanating from easier to solve auxiliary tasks that map local to global connectivity information; and, (ii) adopts an ensemble of Gaussian processes (EGPs) with adaptive weights as nodal embeddings are processed online. Unlike most existing deterministic approaches, the novel approach offers accurate estimates of the unobserved nodal values along with uncertainty quantification that is important especially in safety critical applications. Numerical tests on synthetic and real graph datasets showcase merits of the novel EGP-based Self-SL method.more » « less
-
Facial micro-expressions (MEs) refer to subtle, transient, and involuntary muscle movements expressing a per-son’s true feelings. This paper presents a novel two-stream relational edge-node graph attention network-based approach to classify MEs in a video by selecting the high-intensity frames and edge-node features that can provide valuable information about the relationship between nodes and structural information in a graph structure. The pa-per examines the impact of different edge-node features and their relationships on the graphs. The first step involves extracting high-intensity-emotion frames from the video using optical flow. Second, node feature embeddings are calculated using the node location coordinate features and the patch size information of the optical flow across each node location. Additionally, we obtain the global and local structural similarity score using the jaccard’s similarity score and radial basis function as the edge features. Third, a self-attention graph pooling layer helps to remove the nodes with lower attention scores based on the top-k selection. As the final step, the network employs a two-stream edge-node graph attention network that focuses on finding correlations among the edge and node features, such as landmark coordinates, optical flow, and global and local edge features. A three-frame graph structure is designed to obtain spatio-temporal information. For 3 and 5 expression classes, the results are compared for SMIC and CASME II databases.more » « less
-
Optimizing a black-box function that is expensive to evaluate emerges in a gamut of machine learning and artifcial intelligence applications including drug discovery, policy optimization in robotics, and hyperparameter tuning of learning models to list a few. Bayesian optimization (BO) provides a principled framework to fnd the global optimum of such functions using a limited number of function evaluations. BO relies on a statistical surrogate model to actively select new query points, that is typically captured by a Gaussian process (GP). Unlike most existing approaches that hinge on a single GP surrogate model with a pre-selected kernel function that may confne the expressiveness of the sought function especially under the limited evaluation budget, the present work puts forth a weighted ensemble of GPs as a surrogate model. Building on the advocated Gaussian mixture (GM) posterior, the EGP framework adapts to the most ftted surrogate model as data arrive on-the-fy, offering a richer function space. For the acquisition of next evaluation points, the EGP-based posterior is coupled with an adaptive expected improvement (EI) criterion to balance exploration and exploitation of the search space. Numerical tests on a set of benchmark synthetic functions and two robotic tasks, demonstrate the impressive benefts of the proposed approach.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
