skip to main content


The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 11:00 PM ET on Friday, April 12 until 2:00 AM ET on Saturday, April 13 due to maintenance. We apologize for the inconvenience.

Title: MSNet: Structural Wired Neural Architecture Search for Internet of Things
The prosperity of Internet of Things (IoT) calls for efficient ways of designing extremely compact yet accu- rate DNN models. Both the cell-based neural architec- ture search methods and the recently proposed graph based methods fall short in finding high quality IoT models due to the search flexibility, accuracy density, and node depen- dency limitations. In this paper, we propose a new graph- based neural architecture search methodology MSNAS for crafting highly compact yet accurate models for IoT de- vices. MSNAS supports flexible search space and can ac- cumulate learned knowledge in a meta-graph to increase accuracy density. By adopting structural wiring architec- ture, MSNAS reduces the dependency between nodes, which allows more compact models without sacrificing accuracy. The preliminary experimental results on IoT applications demonstrate that the MSNet crafted by MSNAS outperforms MobileNetV2 and MnasNet by 3.0% in accuracy, with 20% less peak memory consumption and similar Multi-Adds.  more » « less
Award ID(s):
1838024 1756013
Author(s) / Creator(s):
; ; ; ; ;
Date Published:
Journal Name:
Proceedings of ICCV 2019 Neural Architects Workshop
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Inertial navigation provides a small footprint, low-power, and low-cost pathway for localization in GPS-denied environments on extremely resource-constrained Internet-of-Things (IoT) platforms. Traditionally, application-specific heuristics and physics-based kinematic models are used to mitigate the curse of drift in inertial odometry. These techniques, albeit lightweight, fail to handle domain shifts and environmental non-linearities. Recently, deep neural-inertial sequence learning has shown superior odometric resolution in capturing non-linear motion dynamics without human knowledge over heuristic-based methods. These AI-based techniques are data-hungry, suffer from excessive resource usage, and cannot guarantee following the underlying system physics. This paper highlights the unique methods, opportunities, and challenges in porting real-time AI-enhanced inertial navigation algorithms onto IoT platforms. First, we discuss how platform-aware neural architecture search coupled with ultra-lightweight model backbones can yield neural-inertial odometry models that are 31–134 x smaller yet achieve or exceed the localization resolution of state-of-the-art AI-enhanced techniques. The framework can generate models suitable for locating humans, animals, underwater sensors, aerial vehicles, and precision robots. Next, we showcase how techniques from neurosymbolic AI can yield physics-informed and interpretable neural-inertial navigation models. Afterward, we present opportunities for fine-tuning pre-trained odometry models in a new domain with as little as 1 minute of labeled data, while discussing inexpensive data collection and labeling techniques. Finally, we identify several open research challenges that demand careful consideration moving forward. 
    more » « less
  2. null (Ed.)
    As the COVID-19 pandemic evolves, reliable prediction plays an important role in policymaking. The classical infectious disease model SEIR (susceptible-exposed-infectious-recovered) is a compact yet simplistic temporal model. The data-driven machine learning models such as RNN (recurrent neural networks) can suffer in case of limited time series data such as COVID-19. In this paper, we combine SEIR and RNN on a graph structure to develop a hybrid spatiotemporal model to achieve both accuracy and efficiency in training and forecasting. We introduce two features on the graph structure: node feature (local temporal infection trend) and edge feature (geographic neighbor effect). For node feature, we derive a discrete recursion (called I-equation) from SEIR so that gradient descend method applies readily to its optimization. For edge feature, we design an RNN model to capture the neighboring effect and regularize the landscape of loss function so that local minima are effective and robust for prediction. The resulting hybrid model (called IeRNN) improves the prediction accuracy on state-level COVID-19 new case data from the US, out-performing standard temporal models (RNN, SEIR, and ARIMA) in 1-day and 7-day ahead forecasting. Our model accommodates various degrees of reopening and provides potential outcomes for policymakers. 
    more » « less
  3. Graph Neural Networks (GNNs) are becoming increasingly popular for vision-based applications due to their intrinsic capacity in modeling structural and contextual relations between various parts of an image frame. On another front, the rising popularity of deep vision-based applications at the edge has been facilitated by the recent advancements in heterogeneous multi-processor Systems on Chips (MPSoCs) that enable inference under real-time, stringent execution requirements. By extension, GNNs employed for vision-based applications must adhere to the same execution requirements. Yet contrary to typical deep neural networks, the irregular flow of graph learning operations poses a challenge to running GNNs on such heterogeneous MPSoC platforms. In this paper, we propose a novel unifieddesign-mappingapproach for efficient processing of vision GNN workloads on heterogeneous MPSoC platforms. Particularly, we develop MaGNAS, a mapping-aware Graph Neural Architecture Search framework. MaGNAS proposes a GNN architectural design space coupled with prospective mapping options on a heterogeneous SoC to identify model architectures that maximize on-device resource efficiency. To achieve this, MaGNAS employs a two-tier evolutionary search to identify optimalGNNsandmappingpairings that yield the best performance trade-offs. Through designing a supernet derived from the recent Vision GNN (ViG) architecture, we conducted experiments on four (04) state-of-the-art vision datasets using both (i) a real hardware SoC platform (NVIDIA Xavier AGX) and (ii) a performance/cost model simulator for DNN accelerators. Our experimental results demonstrate that MaGNAS is able to provide1.57× latency speedup and is3.38× more energy-efficient for several vision datasets executed on the Xavier MPSoC vs. the GPU-only deployment while sustaining an average0.11%accuracy reduction from the baseline.

    more » « less
  4. In this thesis we propose novel estimation techniques for localization and planning problems, which are key challenges in long-term autonomy. We take inspiration in our methods from non-parametric estimation and use tools such as kernel density estimation, non-linear least-squares optimization, binary masking, and random sampling. We show that these methods, by avoiding explicit parametric models, outperform existing methods that use them. Despite the seeming differences between localization and planning, we demonstrate in this thesis that the problems share core structural similarities. When real or simulation-sampled measurements are expensive, noisy, or high variance, non-parametric estimation techniques give higher-quality results in less time. We first address two localization problems. In order to permit localization with a set of ad hoc-placed radios, we propose an ultra-wideband (UWB) graph realization system to localize the radios. Our system achieves high accuracy and robustness by using kernel density estimation for measurement probability densities, by explicitly modeling antenna delays, and by optimizing this combination with a non-linear least squares formulation. Next, in order to then support robotic navigation, we present a flexible system for simultaneous localization and mapping (SLAM) that combines elements from both traditional dense metric SLAM and topological SLAM, using a binary "masking function" to focus attention. This masking function controls which lidar scans are available for loop closures. We provide several masking functions based on approximate topological class detectors. We then examine planning problems in the final chapter and in the appendix. In order to plan with uncertainty around multiple dynamic agents, we describe Monte-Carlo Policy-Tree Decision Making (MCPTDM), a framework for efficiently computing policies in partially-observable, stochastic, continuous problems. MCPTDM composes a sequence of simpler closed-loop policies and uses marginal action costs and particle repetition to improve cost estimates and sample efficiency by reducing variance. Finally, in the appendix we explore Learned Similarity Monte-Carlo Planning (LSMCP), where we seek to enhance the sample efficiency of partially observable Monte Carlo tree search-based planning by taking advantage of similarities in the final outcomes of similar states and actions. We train a multilayer perceptron to learn a similarity function which we then use to enhance value estimates in the planning. Collectively, we show in this thesis that non-parametric methods promote long-term autonomy by reducing error and increasing robustness across multiple domains. 
    more » « less
  5. Graph Neural Networks (GNNs) have been widely used in various graph-based applications. Recent studies have shown that GNNs are vulnerable to link-level membership inference attacks (LMIA) which can infer whether a given link was included in the training graph of a GNN model. While most of the studies focus on the privacy vulnerability of the links in the entire graph, none have inspected the privacy risk of specific subgroups of links (e.g., links between LGBT users). In this paper, we present the first study of disparity in subgroup vulnerability (DSV) of GNNs against LMIA. First, with extensive empirical evaluation, we demonstrate the existence of non-negligible DSV under various settings of GNN models and input graphs. Second, by both statistical and causal analysis, we identify the difference between three specific graph structural properties of subgroups as one of the underlying reasons for DSV. Among the three properties, the difference between subgroup density has the largest causal effect on DSV. Third, inspired by the causal analysis, we design a new defense mechanism named FairDefense to mitigate DSV while providing protection against LMIA. At a high level, at each iteration of target model training, FairDefense randomizes the membership of edges in the training graph with a given probability, aiming to reduce the gap between the density of different subgroups for DSV mitigation. Our empirical results demonstrate that FairDefense outperforms the existing defense methods in the trade-off between defense and target model accuracy. More importantly, it offers better DSV mitigation.

    more » « less