623509 research outputs found
Sort by
Co-clustering for Federated Recommender System
No full textAs data privacy and security attract increasing attention, Federated Recommender System (FRS) offers a solution that strikes a balance between providing high-quality recommendations and preserving user privacy. However, the presence of statistical heterogeneity in FRS, commonly observed due to personalized decision-making patterns, can pose challenges. To address this issue and maximize the benefit of collaborative filtering (CF) in FRS, it is intuitive to consider clustering clients (users) as well as items into different groups and learning group-specific models. Existing methods either resort to client clustering via user representations-risking privacy leakage, or employ classical clustering strategies on item embeddings or gradients, which we found are plagued by the curse of dimensionality. In this paper, we delve into the inefficiencies of the K-Means method in client grouping, attributing failures due to the high dimensionality as well as data sparsity occurring in FRS, and propose CoFedRec, a novel Co-clustering Federated Recommendation mechanism, to address clients heterogeneity and enhance the collaborative filtering within the federated framework. Specifically, the server initially formulates an item membership from the client-provided item networks. Subsequently, clients are grouped regarding a specific item category picked from the item membership during each communication round, resulting in an intelligently aggregated group model. Meanwhile, to comprehensively capture the global inter-relationships among items, we incorporate an additional supervised contrastive learning term based on the server-side generated item membership into the local training phase for each client. Extensive experiments on four datasets are provided, which verify the effectiveness of the proposed CoFedRec.WWW \u2724: Proceedings of the ACM Web Conference 202
1st-Order Magic: Analysis of Sharpness-Aware Minimization
No full textSharpness-Aware Minimization (SAM) is an optimization technique designed to improve generalization by favoring flatter loss minima. To achieve this, SAM optimizes a modified objective that penalizes sharpness, using computationally efficient approximations. Interestingly, we find that more precise approximations of the proposed SAM objective degrade generalization performance, suggesting that the generalization benefits of SAM are rooted in these approximations rather than in the original intended mechanism. This highlights a gap in our understanding of SAM\u27s effectiveness and calls for further investigation into the role of approximations in optimization.Nalin Tiwary and Siddarth Aananth share equal authorship in this wor
Toward Separating QMA from QCMA with a Classical Oracle
No full textQMA is the class of languages that can be decided by an efficient quantum verifier given a quantum witness, whereas QCMA is the class of such languages where the efficient quantum verifier only is given a classical witness. A challenging fundamental goal in quantum query complexity is to find a classical oracle separation for these classes. In this work, we offer a new approach towards proving such a separation that is qualitatively different than prior work, and show that our approach is sound assuming a natural statistical conjecture which may have other applications to quantum query complexity lower bounds
Fabrication of Ultra-Low-Loss, Dispersion-Engineered Silicon Nitride Photonic Integrated Circuits via Silicon Hardmask Etching
No full textSilicon nitride (SiN) photonic integrated circuits (PICs) have emerged as a versatile platform for a wide range of applications, such as nonlinear optics, narrow-linewidth lasers, and quantum photonics. While thin-film SiN processes have been extensively developed, many nonlinear and quantum optics applications require the use of thick SiN films with engineered dispersion, high mode confinement, and low optical loss. However, high tensile stress in thick SiN films often leads to cracking, making the fabrication challenging to meet these requirements. In this work, we present a robust and reliable fabrication method for ultra-low-loss, dispersion-engineered SiN PICs using amorphous silicon (a-Si) hardmask etching. This approach enables smooth etching of thick SiN waveguides while ensuring long-term storage of crack-free SiN wafers. We achieve intrinsic quality factors () as high as , corresponding to a propagation loss of 1.6 dB/m. The introduction of a-Si hardmask etching and novel crack-isolation trenches offers notable advantages, including high etching selectivity, long-term wafer storage, high yield, and full compatibility with existing well-developed silicon-based semiconductor processes. We demonstrate frequency comb generation in the fabricated microring resonators, showcasing the platform\u27s potential for applications in optical communication, nonlinear optics, metrology, and spectroscopy. This stable and efficient fabrication method offers high performance with significantly reduced fabrication complexity, representing a remarkable advancement toward mass production of SiN PICs for a wide spectrum of applications
Isomorphic gcd-graphs over polynomial rings
No full textGcd-graphs over the ring of integers modulo are a simple and elegant class of integral graphs. The study of these graphs connects multiple areas of mathematics, including graph theory, number theory, and ring theory. In a recent work, inspired by the analogy between number fields and function fields, we define and study gcd-graphs over polynomial rings with coefficients in finite fields. We discover that, in both cases, gcd-graphs share many similar and analogous properties. In this article, we extend this line of research further. Among other topics, we explore an analog of a conjecture of So and a weaker version of Sander-Sander, concerning the conditions under which two gcd-graphs are isomorphic or isospectral. We also provide several constructions showing that, unlike the case over , it is not uncommon for two gcd-graphs over polynomial rings to be isomorphic.Comments are welcome
Site Selective Spontaneous Symmetry Breaking and Partial Order in Kondo Lattices
No full textUsing the combination of a new effective Hamiltonian approach and hybrid Monte-Carlo simulations, we unveil a variety of partially magnetically ordered (PMO) phases in the Kondo lattice model. Our approximation is motivated by two crucial features of the Hamiltonian: (i) formation of Kondo singlets leading to vanishing local magnetic moments, and (ii) spatially correlated nature of the effective single-particle kinetic energy. We discover PMO phases with fractional values , , and of Kondo-screened sites. A common understanding of these states emerges in terms of a non-local ordering mechanism. The concept of site-selective spontaneous symmetry breaking introduced here provides a new general approach to study models of interacting fermions in the intermediate coupling regime.Supplemental material appende
Enhancing Social Robot Navigation with Integrated Motion Prediction and Trajectory Planning in Dynamic Human Environments
No full textNavigating safely in dynamic human environments is crucial for mobile service robots, and social navigation is a key aspect of this process. In this paper, we proposed an integrative approach that combines motion prediction and trajectory planning to enable safe and socially-aware robot navigation. The main idea of the proposed method is to leverage the advantages of Socially Acceptable trajectory prediction and Timed Elastic Band (TEB) by incorporating human interactive information including position, orientation, and motion into the objective function of the TEB algorithms. In addition, we designed social constraints to ensure the safety of robot navigation. The proposed system is evaluated through physical simulation using both quantitative and qualitative metrics, demonstrating its superior performance in avoiding human and dynamic obstacles, thereby ensuring safe navigation. The implementations are open source at: \url{https://github.com/thanhnguyencanh/SGan-TEB.git}In the 24th International Conference on Control, Automation, and Systems (ICCAS 2024), Jeju, Kore
Risk-sensitive control as inference with Rényi divergence
No full textThis paper introduces the risk-sensitive control as inference (RCaI) that extends CaI by using Rényi divergence variational inference. RCaI is shown to be equivalent to log-probability regularized risk-sensitive control, which is an extension of the maximum entropy (MaxEnt) control. We also prove that the risk-sensitive optimal policy can be obtained by solving a soft Bellman equation, which reveals several equivalences between RCaI, MaxEnt control, the optimal posterior for CaI, and linearly-solvable control. Moreover, based on RCaI, we derive the risk-sensitive reinforcement learning (RL) methods: the policy gradient and the soft actor-critic. As the risk-sensitivity parameter vanishes, we recover the risk-neutral CaI and RL, which means that RCaI is a unifying framework. Furthermore, we give another risk-sensitive generalization of the MaxEnt control using Rényi entropy regularization. We show that in both of our extensions, the optimal policies have the same structure even though the derivations are very different.Accepted at NeurIPS 202
Bounding Stochastic Safety: Leveraging Freedman\u27s Inequality with Discrete-Time Control Barrier Functions
No full textWhen deployed in the real world, safe control methods must be robust to unstructured uncertainties such as modeling error and external disturbances. Typical robust safety methods achieve their guarantees by always assuming that the worst-case disturbance will occur. In contrast, this paper utilizes Freedman\u27s inequality in the context of discrete-time control barrier functions (DTCBFs) and c-martingales to provide stronger (less conservative) safety guarantees for stochastic systems. Our approach accounts for the underlying disturbance distribution instead of relying exclusively on its worst-case bound and does not require the barrier function to be upper-bounded, which makes the resulting safety probability bounds more directly useful for intuitive safety constraints such as signed distance. We compare our results with existing safety guarantees, such as input-to-state safety (ISSf) and martingale results that rely on Ville\u27s inequality. When the assumptions for all methods hold, we provide a range of parameters for which our guarantee is stronger. Finally, we present simulation examples, including a bipedal walking robot, that demonstrate the utility and tightness of our safety guarantee.10 pages, 3 figure
Private Attribute Inference from Images with Vision-Language Models
No full textAs large language models (LLMs) become ubiquitous in our daily tasks and digital interactions, associated privacy risks are increasingly in focus. While LLM privacy research has primarily focused on the leakage of model training data, it has recently been shown that LLMs can make accurate privacy-infringing inferences from previously unseen texts. With the rise of vision-language models (VLMs), capable of understanding both images and text, a key question is whether this concern transfers to the previously unexplored domain of benign images posted online. To answer this question, we compile an image dataset with human-annotated labels of the image owner\u27s personal attributes. In order to understand the privacy risks posed by VLMs beyond traditional human attribute recognition, our dataset consists of images where the inferable private attributes do not stem from direct depictions of humans. On this dataset, we evaluate 7 state-of-the-art VLMs, finding that they can infer various personal attributes at up to 77.6% accuracy. Concerningly, we observe that accuracy scales with the general capabilities of the models, implying that future models can be misused as stronger inferential adversaries, establishing an imperative for the development of adequate defenses