arXiv subjects

Su Wang

Publications and source records attributed to Su Wang.

At least 73 records · Page 4Linked to original sources

arXiv2023

Scaling Robot Learning with Semantically Imagined Experience

Recent advances in robot learning have shown promise in enabling robots to perform a variety of manipulation tasks and generalize to novel scenarios. One of the key contributing factors to this progress is the scale of robot data used to train the models. To obtain large-scale datasets, prior approaches have relied on either demonstrations requiring high human involvement or engineering-heavy autonomous data collection schemes, both of which are challenging to scale. To mitigate this issue, we propose an alternative route and leverage text-to-image foundation models widely used in computer vision and natural language processing to obtain meaningful data for robot learning without requiring additional robot data. We term our method Robot Learning with Semantically Imagened Experience (ROSIE). Specifically, we make use of the state of the art text-to-image diffusion models and perform aggressive data augmentation on top of our existing robotic manipulation datasets via inpainting various unseen objects for manipulation, backgrounds, and distractors with text guidance. Through extensive real-world experiments, we show that manipulation policies trained on data augmented this way are able to solve completely unseen tasks with new objects and can behave more robustly w.r.t. novel distractors. In addition, we find that we can improve the robustness and generalization of high-level robot learning tasks such as success detection through training with the diffusion-based data augmentation. The project's website and videos can be found at diffusion-rosie.github.io

cs.RO↗

arXiv2023

How Potent are Evasion Attacks for Poisoning Federated Learning-Based Signal Classifiers?

There has been recent interest in leveraging federated learning (FL) for radio signal classification tasks. In FL, model parameters are periodically communicated from participating devices, training on their own local datasets, to a central server which aggregates them into a global model. While FL has privacy/security advantages due to raw data not leaving the devices, it is still susceptible to several adversarial attacks. In this work, we reveal the susceptibility of FL-based signal classifiers to model poisoning attacks, which compromise the training process despite not observing data transmissions. In this capacity, we develop an attack framework in which compromised FL devices perturb their local datasets using adversarial evasion attacks. As a result, the training process of the global model significantly degrades on in-distribution signals (i.e., signals received over channels with identical distributions at each edge device). We compare our work to previously proposed FL attacks and reveal that as few as one adversarial device operating with a low-powered perturbation under our attack framework can induce the potent model poisoning attack to the global classifier. Moreover, we find that more devices partaking in adversarial poisoning will proportionally degrade the classification performance.

eess.SP↗

arXiv2022

UAV-assisted Online Machine Learning over Multi-Tiered Networks: A Hierarchical Nested Personalized Federated Learning Approach

We investigate training machine learning (ML) models across a set of geo-distributed, resource-constrained clusters of devices through unmanned aerial vehicles (UAV) swarms. The presence of time-varying data heterogeneity and computational resource inadequacy among device clusters motivate four key parts of our methodology: (i) stratified UAV swarms of leader, worker, and coordinator UAVs, (ii) hierarchical nested personalized federated learning (HN-PFL), a distributed ML framework for personalized model training across the worker-leader-core network hierarchy, (iii) cooperative UAV resource pooling to address computational inadequacy of devices by conducting model training among the UAV swarms, and (iv) model/concept drift to model time-varying data distributions. In doing so, we consider both micro (i.e., UAV-level) and macro (i.e., swarm-level) system design. At the micro-level, we propose network-aware HN-PFL, where we distributively orchestrate UAVs inside swarms to optimize energy consumption and ML model performance with performance guarantees. At the macro-level, we focus on swarm trajectory and learning duration design, which we formulate as a sequential decision making problem tackled via deep reinforcement learning. Our simulations demonstrate the improvements achieved by our methodology in terms of ML performance, network resource savings, and swarm trajectory efficiency.

cs.LG↗

arXiv2022

CHES: a space-borne astrometric mission for the detection of habitable planets of the nearby solar-type stars

The Closeby Habitable Exoplanet Survey (CHES) mission is proposed to discover habitable-zone Earth-like planets of the nearby solar-type stars ($\sim 10~\mathrm{pc}$ away from our solar system) via micro-arcsecond relative astrometry. The major scientific objectives of CHES are: to search for Earth Twins or terrestrial planets in habitable zones orbiting 100 FGK nearby stars; further to conduct a comprehensive survey and extensively characterize the nearby planetary systems. The primary payload is a high-quality, low-distortion, high-stability telescope. The optical subsystem is a coaxial three-mirror anastigmat (TMA) with a $1.2 \mathrm{~m}$-aperture, $0.44^{\circ} \times 0.44^{\circ}$ field of view and $500 \mathrm{~nm}-900 \mathrm{~nm}$ working waveband. The camera focal plane is composed of 81 MOSAIC scientific CMOS detectors each with $4 \mathrm{~K} \times 4 \mathrm{~K}$ pixels. The heterodyne laser interferometric calibration technology is employed to ensure micro-arcsecond level (1 $μ$as) relative astrometry precision to meet the requirements for detection of Earth-like planets. CHES satellite operates at the Sun-Earth L2 point and observes the entire target stars for 5 years. CHES will offer the first direct measurements of true masses and inclinations of Earth Twins and super-Earths orbiting our neighbor stars based on micro-arcsecond astrometry from space. This will definitely enhance our understanding of the formation of diverse nearby planetary systems and the emergence of other worlds for solar-type stars, and finally to reflect the evolution of our own solar system.

astro-ph.EP↗

arXiv2022

Less is More: Generating Grounded Navigation Instructions from Landmarks

We study the automatic generation of navigation instructions from 360-degree images captured on indoor routes. Existing generators suffer from poor visual grounding, causing them to rely on language priors and hallucinate objects. Our MARKY-MT5 system addresses this by focusing on visual landmarks; it comprises a first stage landmark detector and a second stage generator -- a multimodal, multilingual, multitask encoder-decoder. To train it, we bootstrap grounded landmark annotations on top of the Room-across-Room (RxR) dataset. Using text parsers, weak supervision from RxR's pose traces, and a multilingual image-text encoder trained on 1.8b images, we identify 971k English, Hindi and Telugu landmark descriptions and ground them to specific regions in panoramas. On Room-to-Room, human wayfinders obtain success rates (SR) of 71% following MARKY-MT5's instructions, just shy of their 75% SR following human instructions -- and well above SRs with other generators. Evaluations on RxR's longer, diverse paths obtain 61-64% SRs on three languages. Generating such high-quality navigation instructions in novel environments is a step towards conversational navigation tools and could facilitate larger-scale training of instruction-following agents.

cs.CV↗

arXiv2022

GenAD: General Representations of Multivariate Time Seriesfor Anomaly Detection

The reliability of wireless base stations in China Mobile is of vital importance, because the cell phone users are connected to the stations and the behaviors of the stations are directly related to user experience. Although the monitoring of the station behaviors can be realized by anomaly detection on multivariate time series, due to complex correlations and various temporal patterns of multivariate series in large-scale stations, building a general unsupervised anomaly detection model with a higher F1-score remains a challenging task. In this paper, we propose a General representation of multivariate time series for Anomaly Detection(GenAD). First, we pre-train a general model on large-scale wireless base stations with self-supervision, which can be easily transferred to a specific station anomaly detection with a small amount of training data. Second, we employ Multi-Correlation Attention and Time-Series Attention to represent the correlations and temporal patterns of the stations. With the above innovations, GenAD increases F1-score by total 9% on real-world datasets in China Mobile, while the performance does not significantly degrade on public datasets with only 10% of the training data.

cs.NI↗

arXiv2022

Coded Caching with Heterogeneous User Profiles

Coded caching utilizes pre-fetching during off-peak hours and multi-casting for delivery in order to balance the traffic load in communication networks. Several works have studied the achievable peak and average rates under different conditions: variable file lengths or popularities, variable cache sizes, decentralized networks, etc. However, very few have considered the possibility of heterogeneous user profiles, despite modern content providers are investing heavily in categorizing users according to their habits and preferences. This paper proposes three coded caching schemes with uncoded pre-fetching for scenarios where end users are grouped into classes with different file demand sets (FDS). One scheme ignores the difference between the classes, another ignores the intersection between them and the third decouples the delivery of files common to all FDS from those unique to a single class. The transmission rates of the three schemes are compared with a lower bound to evaluate their gap to optimality, and with each other to show that each scheme can outperform the other two when certain conditions are met.

cs.IT↗

arXiv2021

The Terrestrial Planet Formation around M Dwarfs: In-situ, Inward Migration or Reversed Migration

Terrestrial planets are commonly observed to orbit M dwarfs with close-in trajectories. In this work, we extensively perform N-body simulations of planetesimal accretion with three models of in-situ, inward migration and reversed migration to explore terrestrial formation in tightly compact systems of M dwarfs. In the simulations, the solid disks are assumed to be 0.01\% of the masses of host stars and spread from 0.01 to 0.5 AU with the surface density profile scaling with $r^{-k}$ according to the observations. Our results show that in-situ scenario may produce $7.77^{+3.23}_{-3.77}$ terrestrial planets with an average mass of $1.23^{+4.01}_{-0.93} \ M_{\oplus}$ around M dwarfs. The number of planets tends to increase as the disk slope is steeper or with a larger stellar mass. Moreover, we show that $2.55^{+1.45}_{-1.55}$ planets with mass of $3.76^{+8.77}_{-3.46} \ M_{\oplus}$ are formed in the systems via inward migration, while $2.85^{+1.15}_{-0.85}$ planets with $3.01^{+13.77}_{-2.71} \ M_{\oplus}$ are yielded under reversed migration. Migration scenarios can also deliver plentiful water from the exterior of ice line to the interior due to more efficient accretion. The simulation outcomes of reversed migration model produce the best matching with observations, being suggestive of a likely mechanism for planetary formation around M dwarfs.

astro-ph.EP↗

arXiv2021

Nii: a Bayesian orbit retrieval code applied to differential astrometry

Here we present an open source Python-based Bayesian orbit retrieval code (Nii) that implements an automatic parallel tempering Markov chain Monte Carlo (APT-MCMC) strategy. Nii provides a module to simulate the observations of a space-based astrometry mission in the search for exoplanets, a signal extraction process for differential astrometric measurements using multiple reference stars, and an orbital parameter retrieval framework using APT-MCMC. We further verify the orbit retrieval ability of the code through two examples corresponding to a single-planet system and a dual-planet system. In both cases, efficient convergence on the posterior probability distribution can be achieved. Although this code specifically focuses on the orbital parameter retrieval problem of differential astrometry, Nii can also be widely used in other Bayesian analysis applications.

astro-ph.IM↗

arXiv2021

threaTrace: Detecting and Tracing Host-based Threats in Node Level Through Provenance Graph Learning

Host-based threats such as Program Attack, Malware Implantation, and Advanced Persistent Threats (APT), are commonly adopted by modern attackers. Recent studies propose leveraging the rich contextual information in data provenance to detect threats in a host. Data provenance is a directed acyclic graph constructed from system audit data. Nodes in a provenance graph represent system entities (e.g., $processes$ and $files$) and edges represent system calls in the direction of information flow. However, previous studies, which extract features of the whole provenance graph, are not sensitive to the small number of threat-related entities and thus result in low performance when hunting stealthy threats. We present threaTrace, an anomaly-based detector that detects host-based threats at system entity level without prior knowledge of attack patterns. We tailor GraphSAGE, an inductive graph neural network, to learn every benign entity's role in a provenance graph. threaTrace is a real-time system, which is scalable of monitoring a long-term running host and capable of detecting host-based intrusion in their early phase. We evaluate threaTrace on three public datasets. The results show that threaTrace outperforms three state-of-the-art host intrusion detection systems.

cs.CR↗

arXiv2021

DeepAID: Interpreting and Improving Deep Learning-based Anomaly Detection in Security Applications

Unsupervised Deep Learning (DL) techniques have been widely used in various security-related anomaly detection applications, owing to the great promise of being able to detect unforeseen threats and superior performance provided by Deep Neural Networks (DNN). However, the lack of interpretability creates key barriers to the adoption of DL models in practice. Unfortunately, existing interpretation approaches are proposed for supervised learning models and/or non-security domains, which are unadaptable for unsupervised DL models and fail to satisfy special requirements in security domains. In this paper, we propose DeepAID, a general framework aiming to (1) interpret DL-based anomaly detection systems in security domains, and (2) improve the practicality of these systems based on the interpretations. We first propose a novel interpretation method for unsupervised DNNs by formulating and solving well-designed optimization problems with special constraints for security domains. Then, we provide several applications based on our Interpreter as well as a model-based extension Distiller to improve security systems by solving domain-specific problems. We apply DeepAID over three types of security-related anomaly detection systems and extensively evaluate our Interpreter with representative prior works. Experimental results show that DeepAID can provide high-quality interpretations for unsupervised DL models while meeting the special requirements of security domains. We also provide several use cases to show that DeepAID can help security operators to understand model decisions, diagnose system mistakes, give feedback to models, and reduce false positives.

cs.CR↗

arXiv2021

Network-Aware Optimization of Distributed Learning for Fog Computing

Fog computing promises to enable machine learning tasks to scale to large amounts of data by distributing processing across connected devices. Two key challenges to achieving this goal are heterogeneity in devices compute resources and topology constraints on which devices can communicate with each other. We address these challenges by developing the first network-aware distributed learning optimization methodology where devices optimally share local data processing and send their learnt parameters to a server for aggregation at certain time intervals. Unlike traditional federated learning frameworks, our method enables devices to offload their data processing tasks to each other, with these decisions determined through a convex data transfer optimization problem that trades off costs associated with devices processing, offloading, and discarding data points. We analytically characterize the optimal data transfer solution for different fog network topologies, showing for example that the value of offloading is approximately linear in the range of computing costs in the network. Our subsequent experiments on testbed datasets we collect confirm that our algorithms are able to improve network resource utilization substantially without sacrificing the accuracy of the learned model. In these experiments, we also study the effect of network dynamics, quantifying the impact of nodes entering or exiting the network on model learning and resource costs.

cs.DC↗

arXiv2021

On the Evaluation of Vision-and-Language Navigation Instructions

Vision-and-Language Navigation wayfinding agents can be enhanced by exploiting automatically generated navigation instructions. However, existing instruction generators have not been comprehensively evaluated, and the automatic evaluation metrics used to develop them have not been validated. Using human wayfinders, we show that these generators perform on par with or only slightly better than a template-based generator and far worse than human instructors. Furthermore, we discover that BLEU, ROUGE, METEOR and CIDEr are ineffective for evaluating grounded navigation instructions. To improve instruction evaluation, we propose an instruction-trajectory compatibility model that operates without reference instructions. Our model shows the highest correlation with human wayfinding outcomes when scoring individual instructions. For ranking instruction generation systems, if reference instructions are available we recommend using SPICE.

cs.AI↗

arXiv2021

Device Sampling for Heterogeneous Federated Learning: Theory, Algorithms, and Implementation

The conventional federated learning (FedL) architecture distributes machine learning (ML) across worker devices by having them train local models that are periodically aggregated by a server. FedL ignores two important characteristics of contemporary wireless networks, however: (i) the network may contain heterogeneous communication/computation resources, while (ii) there may be significant overlaps in devices' local data distributions. In this work, we develop a novel optimization methodology that jointly accounts for these factors via intelligent device sampling complemented by device-to-device (D2D) offloading. Our optimization aims to select the best combination of sampled nodes and data offloading configuration to maximize FedL training accuracy subject to realistic constraints on the network topology and device capabilities. Theoretical analysis of the D2D offloading subproblem leads to new FedL convergence bounds and an efficient sequential convex optimizer. Using this result, we develop a sampling methodology based on graph convolutional networks (GCNs) which learns the relationship between network attributes, sampled nodes, and resulting offloading that maximizes FedL accuracy. Through evaluation on real-world datasets and network measurements from our IoT testbed, we find that our methodology while sampling less than 5% of all devices outperforms conventional FedL substantially both in terms of trained model accuracy and required resource utilization.

cs.NI↗

arXiv2020

Departure from the Exact Location of Mean Motion Resonances Induced by the Gas Disk in the Systems Observed by Kepler

The statistical results of transiting planets show that there are two peaks around 1.5 and 2.0 in the distribution of orbital period ratios. A large number of planet pairs are found near the exact location of mean motion resonances (MMRs). In this work, we find out that the depletion and structures of gas disk play crucial roles in driving planet pairs out of exact location of MMRs. Under such scenario, planet pairs are trapped into exact MMRs during orbital migration firstly and keep migrating in a same pace. The eccentricities can be excited. Due to the existence of gas disk, eccentricities can be damped leading to the change of orbital period. It will make planet pairs depart from the exact location of MMRs. With depletion timescales larger than 1 Myr, near MMRs configurations are formed easily. Planet pairs have higher possibilities to escape from MMRs with higher disk aspect ratio. Additionally, with weaker corotation torque, planet pairs can depart farther from exact location of MMRs. The final location of the innermost planets in systems are directly related to the transition radius from optically thick region to inner optically thin disk. While the transition radius is smaller than 0.2 AU at the late stage of star evolution process, the innermost planets can reach around 10 days. Our formation scenario is a possible mechanism to explain the formation of near MMRs configuration with the innermost planet farther than 0.1 AU.

astro-ph.EP↗

arXiv2020

Narrative Interpolation for Generating and Understanding Stories

We propose a method for controlled narrative/story generation where we are able to guide the model to produce coherent narratives with user-specified target endings by interpolation: for example, we are told that Jim went hiking and at the end Jim needed to be rescued, and we want the model to incrementally generate steps along the way. The core of our method is an interpolation model based on GPT-2 which conditions on a previous sentence and a next sentence in a narrative and fills in the gap. Additionally, a reranker helps control for coherence of the generated text. With human evaluation, we show that ending-guided generation results in narratives which are coherent, faithful to the given ending guide, and require less manual effort on the part of the human guide writer than past approaches.

cs.CL↗

arXiv2020

The Definition and Long-term Variations of Beijing Blue Days

The phenomenon of Beijing Blue days first occurs on June 11, 2015, and become a hot society topic in a short time. Thus, Beijing's blue day is not only what ordinary people desire but national needs. In this work, using the data of daily average meteorological observations during 1980 and 2014, we select three standards for defining Beijing blue day (BBI), which contain no rain, clear sky, and great dry visibility, and the accuracy is 73.4%.Our study finds that Beijing shares the most significant acceleration in the BTH region, with a growth rate of 2.66 d/year. The greatest seasonal average and the maximum acceleration rate appear in winter for total BBI and continuous BBI days, the least is in summer, while Beijing blue appears most frequently in January. BBI is more related to relative humidity and rainless day on an annual scale, but different in different seasons.

physics.data-an↗

arXiv2020

Near Mean Motion Resonance of Terrestrial Planet Pair Induced by Giant Planet: Application to Kepler-68 System

In this work, we investigate configuration formation of two inner terrestrial planets near mean motion resonance (MMRs) induced by the perturbation of a distant gas-giant for the Kepler-68 system, by conducting thousands of numerical simulations. The results show that the formation of terrestrial planets is relevant to the speed of Type I migration, the mass of planets, and the existence of giant planet. The mass and eccentricity of the giant planet may play a crucial role in shaping the final configuration of the system. The inner planet pair can be trapped in 5:3 or 7:4 MMRs if the giant planet revolves the central star with an eccentric orbit, which is similar to the observed configuration of Kepler-68. Moreover, we find that the eccentricity of the middle planet can be excited to roughly 0.2 if the giant planet is more massive than 5 $M_J$,otherwise the terrestrial planets are inclined to remain near-circular orbits. Our study may provide a likely formation scenario for the planetary systems that harbor several terrestrial planets near MMRs inside and one gas-giant exterior to them.

astro-ph.EP↗