Spiking Neural Networks (SNNs) and data from Dynamic Vision Sensors (DVSs) offer energy-efficient solutions for edge devices with limited battery life. The input latencies of event data to SNNs are critical for energy savings, and reducing these latencies through configurable parameters is essential. However, security concerns, particularly adversarial attacks on SNNs, are increasingly significant. While most existing research primarily focuses on attacking event sequences, which may not always be accessible, attacks on raw event streams with configurable latencies remain underexplored due to challenges such as extreme sparsity and the complexity of discrete optimization. This paper proposes a novel adversarial attack method on raw event streams with configurable latencies. To address sparsity and discrete optimization, we smooth the optimization by converting binary spikes into continuous values. Furthermore, we introduce an adaptively stochastic strategy for sampling attacking latencies. We apply regularization terms to maintain sparsity and ensure adversarial samples resemble raw event streams while approximating the target label. Extensive experiments on datasets such as N-MNIST, CIFAR10-DVS, N-Caltech-101, and Gesture-DVS demonstrate that our method consistently outperforms existing approaches, achieving higher attack success rates (ASR) across various latencies. Ablation studies validate the effectiveness of our contributions and highlight the impact of latency on the generation of adversarial samples.

In the field of computer vision, event-based Dynamic Vision Sensors (DVSs) have emerged as a significant complement to traditional pixel-based imaging due to their low power consumption and high temporal resolution. These sensors, particularly when combined with Spiking Neural Networks (SNNs), offer a promising direction for energy-efficient and fast-reacting vision systems. Typically, DVS data are converted into grid-based formats for processing with SNNs, with this transformation process often being an opaque step in the pipeline. As a result, the grid representation becomes an intermediate yet inaccessible stage during the implementation of attacks, highlighting the importance of attacking raw event data. Existing attack methodologies predominantly target grid-based representations, hindered by the complexity of three-valued optimization and the broad optimization space associated with raw event data. Our study addresses this gap by introducing a novel adversarial attack approach that directly targets raw event data. We tackle the inherent challenges of three-valued optimization and the need to preserve data sparsity through a strategic amalgamation of methods: 1) Treating Discrete Event Values as Probabilistic Samples: This allows for continuous optimization by considering discrete event values as probabilistic space samples. 2) Focusing on Specific Event Positions: We prioritize specific event positions that merge original data with additional target label data, enhancing attack precision. 3) Employing a Sparsity Norm: To retain the original data's sparsity, a sparsity norm is utilized, ensuring the adversarial data's comparability. Our empirical findings demonstrate the effectiveness of our combined approach, achieving noteworthy success in targeted attacks and highlighting vulnerabilities in models based on raw event data.

Spiking neural networks are becoming increasingly popular for their low energy requirement in real-world tasks with accuracy comparable to the traditional ANNs. SNN training algorithms face the loss of gradient information and non-differentiability due to the Heaviside function in minimizing the model loss over model parameters. To circumvent the problem surrogate method uses a differentiable approximation of the Heaviside in the backward pass, while the forward pass uses the Heaviside as the spiking function. We propose to use the zeroth order technique at the neuron level to resolve this dichotomy and use it within the automatic differentiation tool. As a result, we establish a theoretical connection between the proposed local zeroth-order technique and the existing surrogate methods and vice-versa. The proposed method naturally lends itself to energy-efficient training of SNNs on GPUs. Experimental results with neuromorphic datasets show that such implementation requires less than 1 percent neurons to be active in the backward pass, resulting in a 100x speed-up in the backward computation time. Our method offers better generalization compared to the state-of-the-art energy-efficient technique while maintaining similar efficiency.

On-device learning faces memory constraints when optimizing or fine-tuning on edge devices with limited resources. Current techniques for training deep models on edge devices rely heavily on backpropagation. However, its high memory usage calls for a reassessment of its dominance. In this paper, we propose forward gradient descent (FGD) as a potential solution to overcome the memory capacity limitation in on-device learning. However, FGD's dependencies across layers hinder parallel computation and can lead to inefficient resource utilization. To mitigate this limitation, we propose AsyncFGD, an asynchronous framework that decouples dependencies, utilizes module-wise stale parameters, and maximizes parallel computation. We demonstrate its convergence to critical points through rigorous theoretical analysis. Empirical evaluations conducted on NVIDIA's AGX Orin, a popular embedded device, show that AsyncFGD reduces memory consumption and enhances hardware efficiency, offering a novel approach to on-device learning.

Nowadays self-paced learning (SPL) is an important machine learning paradigm that mimics the cognitive process of humans and animals. The SPL regime involves a self-paced regularizer and a gradually increasing age parameter, which plays a key role in SPL but where to optimally terminate this process is still non-trivial to determine. A natural idea is to compute the solution path w.r.t. age parameter (i.e., age-path). However, current age-path algorithms are either limited to the simplest regularizer, or lack solid theoretical understanding as well as computational efficiency. To address this challenge, we propose a novel Generalized Age-path Algorithm (GAGA) for SPL with various self-paced regularizers based on ordinary differential equations (ODEs) and sets control, which can learn the entire solution spectrum w.r.t. a range of age parameters. To the best of our knowledge, GAGA is the first exact path-following algorithm tackling the age-path for general self-paced regularizer. Finally the algorithmic steps of classic SVM and Lasso are described in detail. We demonstrate the performance of GAGA on real-world datasets, and find considerable speedup between our algorithm and competing baselines.

Link prediction in dynamic networks is a well studied topic. Yet until recently, validation of algorithms has been hampered by limitations in the size and realism of empirical datasets. In this work, we seek to revisit and reassess the value and accuracy of prediction methods, by leveraging our access to several large, detailed traces of dynamics in online social networks (Facebook, Renren, YouTube). Our goals are to understand the absolute and comparative accuracy of existing prediction algorithms, and to develop techniques to improve them using insights from analysis of network dynamics. We implement and evaluate 18 link prediction algorithms, labeled as either "metric-based" (those that predict potential links using a single similarity or proximity metric) or "classification-based" (those that use machine learning classifiers with multiple metrics as input features). Despite poor performance in absolute terms, SVM classifiers consistently perform the best across all our traces. Its accuracy is occasionally matched by metric-based algorithms, but never consistently across datasets. Finally, we use observations of network dynamics to build "filters" that dramatically reduce the search space for link candidates. Augmenting current algorithms with our filters dramatically improves prediction accuracy across all traces and algorithms.

Rack-scale computers are dense clusters with hundreds of micro-servers per rack. Designed for data center workloads, they can have significant power, cost and performance benefits over current racks. The rack network can be distributed, with small packet switches embedded on each processor as part of a system-on-chip (SoC) design. Ingress/egress traffic is forwarded by SoCs that have direct uplinks to the data center. Such fabrics are not fully provisioned and the chosen topology and uplink placement impacts performance for different workloads.XFabric is a rack-scale network that reconfigures the topology and uplink placement using a circuit-switched physical layer over which SoCs perform packet switching. To satisfy tight power and space requirements in the rack, XFabric does not use a single large circuit switch, instead relying on a set of independent smaller circuit switches. This introduces partial reconfigurability, as some ports in the rack cannot be connected by a circuit. XFabric optimizes the physical topology and manages uplinks, efficiently coping with partial reconfigurability. It significantly outperforms static topologies and has a performance similar to fully reconfigurable fabrics. We demonstrate the benefits of XFabric using flow-based simulations and a prototype built with electrical cross-point switch ASICs.

Analyzing and modeling social network dynamics are key to accurately predicting resource needs and system behavior in online social networks. The presence of statistical scaling properties, that is, self-similarity, is critical for determining how to model network dynamics. In this work, we study the role that self-similarity scaling plays in a social network edge creation (that is, links created between users) process, through analysis of two detailed, time-stamped traces, a 199 million edge trace over 2 years in the Renren social network, and 876K interactions in a 4-year trace of Facebook. Using wavelet-based analysis, we find that the edge creation process in both networks is consistent with self-similarity scaling, once we account for periodic user activity that makes edge creation process non-stationary. Using these findings, we build a complete model of social network dynamics that combines temporal and spatial components. Specifically, the temporal behavior of our model reflects self-similar scaling properties, and accounts for certain deterministic non-stationary features. The spatial side accounts for observed long-term graph properties, such as graph distance shrinkage and local declustering. We validate our model against network dynamics in Renren and Facebook datasets, and show that it succeeds in producing desired properties in both temporal patterns and graph structural features.

From network topologies to online social networks, many of today's most sensitive datasets are captured in large graphs. A significant challenge facing owners of these datasets is how to share sensitive graphs with collaborators and authorized users, e.g. network topologies with network equipment vendors or Facebook's social graphs with academic collaborators. Current tools can provide limited node or edge privacy, but require modifications to the graph that significantly reduce its utility. In this work, we propose a new alternative in the form of graph watermarks. Graph watermarks are small graphs tailor-made for a given graph dataset, a secure graph key, and a secure user key. To share a sensitive graph G with a collaborator C, the owner generates a watermark graph W using G, the graph key, and C's key as input, and embeds W into G to form G'. If G' is leaked by C,its owner can reliably determine if the watermark W generated for C does in fact reside inside G', thereby proving C is responsible for the leak. Graph watermarks serve both as a deterrent against data leakage and a method of recourse after a leak. We provide robust schemes for creating, embedding and extracting watermarks, and use analysis and experiments on large, real graphs to show that they are unique and difficult to forge. We study the robustness of graph watermarks against both single and powerful colluding attacker models, then propose and empirically evaluate mechanisms to dramatically improve resilience.

From network topologies to online social networks, many of today's most sensitive datasets are captured in large graphs. A significant challenge facing the data owners is how to share sensitive graphs with collaborators or authorized users, e.g. ISP's network topology graphs with a third party networking equipment vendor. Current tools can provide limited node or edge privacy, but significantly modify the graph reducing its utility. In this work, we propose a new alternative in the form of graph watermarks . Graph watermarks are small graphs tailor-made for a given graph dataset, a secure graph key, and a secure user key. To share a sensitive graph G with a collaborator C , the owner generates a watermark graph W using G , the graph key, and C 's key as input, and embeds W into G to form G' . If G' is leaked by C , its owner can reliably determine if the watermark W generated for C does in fact reside inside G' , thereby proving C is responsible for the leak. Graph watermarks serve both as a deterrent against data leakage and a method of recourse after a leak. We provide robust schemes for embedding and extracting watermarks, and use analysis and experiments on large real graphs to show that they are unique and difficult to forge. We study the robustness of graph watermarks against both single and powerful colluding attacker models, then propose and evaluate mechanisms to dramatically improve resilience.

Today, numerous models and metrics are available to capture and characterize static properties of online social networks. When it comes to understanding their dynamics and evolution, however, research offers little interms of metrics or models. Current metrics are limited to logical time clocks, and unable to capture interactions with external factors that rely on physical time clocks. In this paper, our goal is to take initial steps towards building a set of metrics for characterizing social network dynamics based on physical time. We focus our attention on two metrics that capture the "eagerness" of users in building social structure. More specifically, we propose metrics of link delay and triadic closure delay, two metrics that capture the time delay between when a link or triadic closure is possible,and when they actually instantiate in the trace. Considered over time or across traces, the value of these metrics can provide insight on the speed at which users act in building and extending their social neighborhoods. We apply these metrics to two real traces of social network dynamics from the Renren and Facebook networks. We show that these metrics are generally consistent across networks, but their differences reveal interesting properties of each system. We argue that they can be attributed to factors such as network maturity, environmental and social contexts, and services offered by network provider, all factors independent of the network topology and captured by our proposed metrics. Finally, we find that triadic closure delays capture the ease of neighbor discovery in social networks, and can be strongly influenced by friend recommendation systems.

Analysis of large graphs is critical to the ongoing growth of search engines and social networks. One class of queries centers around node affinity, often quantified by random-walk distances between node pairs, including hitting time, commute time, and personalized PageRank (PPR). Despite the potential of these "metrics," they are rarely, if ever, used in practice, largely due to extremely high computational costs.In this paper, we investigate methods to scalably and efficiently compute random-walk distances, by "embedding" graphs and distances into points and distances in geometric coordinate spaces. We show that while existing graph coordinate systems (GCS) can accurately estimate shortest path distances, they produce significant errors when embedding random-walk distances. Based on our observations, we propose a new graph embedding system that explicitly accounts for per-node graph properties that affect random walk. Extensive experiments on a range of graphs show that our new approach can accurately estimate both symmetric and asymmetric random-walk distances. Once a graph is embedded, our system can answer queries between any two nodes in 8 microseconds, orders of magnitude faster than existing methods. Finally, we show that our system produces estimates that can replace ground truth in applications with minimal impact on application output.

Measurement studies of online social networks show that all social links are not equal, and the strength of each link is best characterized by the frequency of interactions between the linked users.To date, few studies have been able to examine detailed interaction data over time, and none have studied the problem of modeling user interactions. This paper proposes a generative model of social interactions that captures the inherently heterogeneous strengths of social links, thus having broad implications on the design of social network algorithms such as friend recommendation, information diffusion and viral marketing.

Mobile networking researchers have long searched for large-scale, fine-grained traces of human movement, which have remained elusive for both privacy and logistical reasons. Recently, researchers have begun to focus on geosocial mobility traces, e.g. Foursquare checkin traces, because of their availability and scale. But are we conceding correctness in our zeal for data? In this paper, we take initial steps towards quantifying the value of geosocial datasets using a large ground truth dataset gathered from a user study. By comparing GPS traces against Foursquare checkins, we find that a large portion of visited locations is missing from checkins, and most checkin events are either forged or superfluous events. We characterize extraneous checkins, describe possible techniques for their detection, and show that both extraneous and missing checkins introduce significant errors into applications driven by these traces.

Data confidentiality policies at major social network providers have severely limited researchers' access to large-scale datasets. The biggest impact has been on the study of network dynamics, where researchers have studied citation graphs and content-sharing networks, but few have analyzed detailed dynamics in the massive social networks that dominate the web today. In this paper, we present results of analyzing detailed dynamics in a large Chinese social network, covering a period of 2 years when the network grew from its first user to 19 million users and 199 million edges. Rather than validate a single model of network dynamics, we analyze dynamics at different granularities (per-user, per-community, and network-wide) to determine how much, if any, users are influenced by dynamics processes at different scales. We observe independent predictable processes at each level, and find that the growth of communities has moderate and sustained impact on users. In contrast, we find that significant events such as network merge events have a strong but short-lived impact on users, and they are quickly eclipsed by the continuous arrival of new users.

Popular Internet services in recent years have shown that remarkable things can be achieved by harnessing the power of the masses using crowd-sourcing systems. However, crowd-sourcing systems can also pose a real challenge to existing security mechanisms deployed to protect Internet services. Many of these security techniques rely on the assumption that malicious activity is generated automatically by automated programs. Thus they would perform poorly or be easily bypassed when attacks are generated by real users working in a crowd-sourcing system. Through measurements, we have found surprising evidence showing that not only do malicious crowd-sourcing systems exist, but they are rapidly growing in both user base and total revenue. We describe in this paper a significant effort to study and understand these "crowdturfing" systems in today's Internet. We use detailed crawls to extract data about the size and operational structure of these crowdturfing systems. We analyze details of campaigns offered and performed in these sites, and evaluate their end-to-end effectiveness by running active, benign campaigns of our own. Finally, we study and compare the source of workers on crowdturfing sites in different countries. Our results suggest that campaigns on these systems are highly effective at reaching users, and their continuing growth poses a concrete threat to online communities both in the US and elsewhere.

Network Coordinate Systems (NCS) allow accurate latency estimation with low measurement overhead. However, NCS are vulnerable to attacks, which makes security issue a crucial concern in their large-scale deployment. This paper proposes Sniper, a social relationship based defence mechanism for securing NCS. In contrast to prior proposals that allow attacks to occur and then detect them, Sniper can filter out potential malicious nodes before they mount attack. Extensive simulations demonstrate that Sniper can protect NCS against various forms of attacks efficiently and that Sniper can tolerate a large fraction of malicious nodes.

Graph analysis is a critical component of applications such as online social networks, protein interactions in biological networks, and Internet traffic analysis. The arrival of massive graphs with hundreds of millions of nodes, e.g. social graphs, presents a unique challenge to graph analysis applications. Most of these applications rely on computing distances between node pairs, which for large graphs can take minutes to compute using traditional algorithms such as breadth-first-search (BFS). In this paper, we study ways to enable scalable graph processing on today's massive graphs. We explore the design space of graph coordinate systems, a new approach that accurately approximates node distances in constant time by embedding graphs into coordinate spaces. We show that a hyperbolic embedding produces relatively low distortion error, and propose Rigel, a hyperbolic graph coordinate system that lends itself to efficient parallelization across a compute cluster. Rigel produces significantly more accurate results than prior systems, and is naturally parallelizable across compute clusters, allowing it to provide accurate results for graphs up to 43 million nodes. Finally, we show that Rigel's functionality can be easily extended to locate (near-) shortest paths between node pairs. After a one- time preprocessing cost, Rigel answers node-distance queries in 10's of microseconds, and also produces shortest path results up to 18 times faster than prior shortest-path systems with similar levels of accuracy.

Analysis of large networks is a critical component of many of today's application environments, including online social networks, protein interactions in biological networks, and Internet traffic analysis. The arrival of massive network graphs with hundreds of millions of nodes, e.g. social graphs, presents a unique challenge to graph analysis applications. Most of these applications rely on computing distances between node pairs, which for large graphs can take minutes to compute using traditional algorithms such as breadth-first-search (BFS). In this paper, we study ways to enable scalable graph processing for today's massive networks. We explore the design space of graph coordinate systems, a new approach that accurately approximates node distances in constant time by embedding graphs into coordinate spaces. We show that a hyperbolic embedding produces relatively low distortion error, and propose Rigel, a hyperbolic graph coordinate system that lends itself to efficient parallelization across a compute cluster. Rigel produces significantly more accurate results than prior systems, and is naturally parallelizable across compute clusters, allowing it to provide accurate results for graphs up to 43 million nodes. Finally, we show that Rigel's functionality can be easily extended to locate (near-) shortest paths between node pairs. After a onetime preprocessing cost, Rigel answers node-distance queries in 10's of microseconds, and also produces shortest path results up to 18 times faster than prior shortest-path systems with similar levels of accuracy.