Sina Sajadmanesh

Sina Sajadmanesh, Ali Shahin Shamsabadi, Aurélien Bellet et al.

In this paper, we study the problem of learning Graph Neural Networks (GNNs) with Differential Privacy (DP). We propose a novel differentially private GNN based on Aggregation Perturbation (GAP), which adds stochastic noise to the GNN's aggregation function to statistically obfuscate the presence of a single edge (edge-level privacy) or a single node and all its adjacent edges (node-level privacy). Tailored to the specifics of private learning, GAP's new architecture is composed of three separate modules: (i) the encoder module, where we learn private node embeddings without relying on the edge information; (ii) the aggregation module, where we compute noisy aggregated node embeddings based on the graph structure; and (iii) the classification module, where we train a neural network on the private aggregations for node classification without further querying the graph edges. GAP's major advantage over previous approaches is that it can benefit from multi-hop neighborhood aggregations, and guarantees both edge-level and node-level DP not only for training, but also at inference with no additional costs beyond the training's privacy budget. We analyze GAP's formal privacy guarantees using Rényi DP and conduct empirical experiments over three real-world graph datasets. We demonstrate that GAP offers significantly better accuracy-privacy trade-offs than state-of-the-art DP-GNN approaches and naive MLP-based baselines. Our code is publicly available at https://github.com/sisaman/GAP.

Sina Sajadmanesh, Daniel Gatica-Perez

Graph Neural Networks (GNNs) have become a popular tool for learning on graphs, but their widespread use raises privacy concerns as graph data can contain personal or sensitive information. Differentially private GNN models have been recently proposed to preserve privacy while still allowing for effective learning over graph-structured datasets. However, achieving an ideal balance between accuracy and privacy in GNNs remains challenging due to the intrinsic structural connectivity of graphs. In this paper, we propose a new differentially private GNN called ProGAP that uses a progressive training scheme to improve such accuracy-privacy trade-offs. Combined with the aggregation perturbation technique to ensure differential privacy, ProGAP splits a GNN into a sequence of overlapping submodels that are trained progressively, expanding from the first submodel to the complete model. Specifically, each submodel is trained over the privately aggregated node embeddings learned and cached by the previous submodels, leading to an increased expressive power compared to previous approaches while limiting the incurred privacy costs. We formally prove that ProGAP ensures edge-level and node-level privacy guarantees for both training and inference stages, and evaluate its performance on benchmark graph datasets. Experimental results demonstrate that ProGAP can achieve up to 5-10% higher accuracy than existing state-of-the-art differentially private GNNs. Our code is available at https://github.com/sisaman/ProGAP.

Jiabo Huang, Zhizhong Li, Sina Sajadmanesh et al.

Deploying vision-language models (VLMs) in resource-constrained settings demands low latency and high throughput, yet existing compact VLMs often fall short of the inference speedups their smaller parameter counts suggest. To explain this discrepancy, we conduct an empirical end-to-end efficiency analysis and systematically profile inference to identify the dominant bottlenecks. Based on these findings, we develop optimization recipes tailored to compact VLMs that substantially reduce latency while preserving accuracy. These techniques cut time to first token (TTFT) by 53% on InternVL3-2B and by 93% on SmolVLM-256M. Our recipes are broadly applicable across both VLM architectures and common serving frameworks, providing practical guidance for building efficient VLM systems. Beyond efficiency, we study how to extend compact VLMs with structured perception outputs and introduce the resulting model family, ArgusVLM. Across diverse benchmarks, ArgusVLM achieves strong performance while maintaining a compact and efficient design.

Zhizhong Li, Sina Sajadmanesh, Jingtao Li et al.

Low-rank adaptation (LoRA) has been widely adopted as a parameter-efficient technique for fine-tuning large-scale pre-trained models. However, it still lags behind full fine-tuning in performance, partly due to its insufficient exploitation of the geometric structure underlying low-rank manifolds. In this paper, we propose a geometry-aware extension of LoRA that uses a three-factor decomposition $U\!SV^\top$. Analogous to the structure of singular value decomposition (SVD), it separates the adapter's input and output subspaces, $V$ and $U$, from the scaling factor $S$. Our method constrains $U$ and $V$ to lie on the Stiefel manifold, ensuring their orthonormality throughout the training. To optimize on the Stiefel manifold, we employ a flexible and modular geometric optimization design that converts any Euclidean optimizer to a Riemannian one. It enables efficient subspace learning while remaining compatible with existing fine-tuning pipelines. Empirical results across a wide range of downstream tasks, including commonsense reasoning, math and code generation, image classification, and image generation, demonstrate the superior performance of our approach against the recent state-of-the-art variants of LoRA. Code is available at https://github.com/SonyResearch/stella.

David Schneider, Sina Sajadmanesh, Vikash Sehwag et al. · princeton

Privacy-preserving computer vision is an important emerging problem in machine learning and artificial intelligence. Prevalent methods tackling this problem use differential privacy (DP) or obfuscation techniques to protect the privacy of individuals. In both cases, the utility of the trained model is sacrificed heavily in this process. In this work, we present an anonymization pipeline that replaces sensitive human subjects in video datasets with synthetic avatars within context, employing a combined rendering and stable diffusion-based strategy. Additionally we propose masked differential privacy ({MaskDP}) to protect non-anonymized but privacy sensitive background information. MaskDP allows for controlling sensitive regions where differential privacy is applied, in contrast to applying DP on the entire input. This combined methodology provides strong privacy protection while minimizing the usual performance penalty of privacy preserving methods. Experiments on multiple challenging action recognition datasets demonstrate that our proposed techniques result in better utility-privacy trade-offs compared to standard differentially private training in the especially demanding $ε<1$ regime.

Sina Sajadmanesh, Daniel Gatica-Perez

Graph Neural Networks (GNNs) have demonstrated superior performance in learning node representations for various graph inference tasks. However, learning over graph data can raise privacy concerns when nodes represent people or human-related variables that involve sensitive or personal information. While numerous techniques have been proposed for privacy-preserving deep learning over non-relational data, there is less work addressing the privacy issues pertained to applying deep learning algorithms on graphs. In this paper, we study the problem of node data privacy, where graph nodes have potentially sensitive data that is kept private, but they could be beneficial for a central server for training a GNN over the graph. To address this problem, we develop a privacy-preserving, architecture-agnostic GNN learning algorithm with formal privacy guarantees based on Local Differential Privacy (LDP). Specifically, we propose an LDP encoder and an unbiased rectifier, by which the server can communicate with the graph nodes to privately collect their data and approximate the GNN's first layer. To further reduce the effect of the injected noise, we propose to prepend a simple graph convolution layer, called KProp, which is based on the multi-hop aggregation of the nodes' features acting as a denoising mechanism. Finally, we propose a robust training framework, in which we benefit from KProp's denoising capability to increase the accuracy of inference in the presence of noisy labels. Extensive experiments conducted over real-world datasets demonstrate that our method can maintain a satisfying level of accuracy with low privacy loss.

Sina Sajadmanesh, Sogol Bazargani, Jiawei Zhang et al.

Online social networks, World Wide Web, media and technological networks, and other types of so-called information networks are ubiquitous nowadays. These information networks are inherently heterogeneous and dynamic. They are heterogeneous as they consist of multi-typed objects and relations, and they are dynamic as they are constantly evolving over time. One of the challenging issues in such heterogeneous and dynamic environments is to forecast those relationships in the network that will appear in the future. In this paper, we try to solve the problem of continuous-time relationship prediction in dynamic and heterogeneous information networks. This implies predicting the time it takes for a relationship to appear in the future, given its features that have been extracted by considering both heterogeneity and temporal dynamics of the underlying network. To this end, we first introduce a feature extraction framework that combines the power of meta-path-based modeling and recurrent neural networks to effectively extract features suitable for relationship prediction regarding heterogeneity and dynamicity of the networks. Next, we propose a supervised non-parametric approach, called Non-Parametric Generalized Linear Model (NP-GLM), which infers the hidden underlying probability distribution of the relationship building time given its features. We then present a learning algorithm to train NP-GLM and an inference method to answer time-related queries. Extensive experiments conducted on synthetic data and three real-world datasets, namely Delicious, MovieLens, and DBLP, demonstrate the effectiveness of NP-GLM in solving continuous-time relationship prediction problem vis-a-vis competitive baselines

Sina Sajadmanesh, Jiawei Zhang, Hamid R. Rabiee

In this paper, we try to solve the problem of temporal link prediction in information networks. This implies predicting the time it takes for a link to appear in the future, given its features that have been extracted at the current network snapshot. To this end, we introduce a probabilistic non-parametric approach, called "Non-Parametric Generalized Linear Model" (NP-GLM), which infers the hidden underlying probability distribution of the link advent time given its features. We then present a learning algorithm for NP-GLM and an inference method to answer time-related queries. Extensive experiments conducted on both synthetic data and real-world Sina Weibo social network demonstrate the effectiveness of NP-GLM in solving temporal link prediction problem vis-a-vis competitive baselines.

Seyed Ali Osia, Ali Shahin Shamsabadi, Sina Sajadmanesh et al.

Internet of Things (IoT) devices and applications are being deployed in our homes and workplaces. These devices often rely on continuous data collection to feed machine learning models. However, this approach introduces several privacy and efficiency challenges, as the service operator can perform unwanted inferences on the available data. Recently, advances in edge processing have paved the way for more efficient, and private, data processing at the source for simple tasks and lighter models, though they remain a challenge for larger, and more complicated models. In this paper, we present a hybrid approach for breaking down large, complex deep neural networks for cooperative, privacy-preserving analytics. To this end, instead of performing the whole operation on the cloud, we let an IoT device to run the initial layers of the neural network, and then send the output to the cloud to feed the remaining layers and produce the final result. In order to ensure that the user's device contains no extra information except what is necessary for the main task and preventing any secondary inference on the data, we introduce Siamese fine-tuning. We evaluate the privacy benefits of this approach based on the information exposed to the cloud service. We also assess the local inference cost of different layers on a modern handset. Our evaluations show that by using Siamese fine-tuning and at a small processing cost, we can greatly reduce the level of unnecessary, potentially sensitive information in the personal data, and thus achieving the desired trade-off between utility, privacy, and performance.

Sina Sajadmanesh, Sina Jafarzadeh, Seyed Ali Osia et al.

Food and nutrition occupy an increasingly prevalent space on the web, and dishes and recipes shared online provide an invaluable mirror into culinary cultures and attitudes around the world. More specifically, ingredients, flavors, and nutrition information become strong signals of the taste preferences of individuals and civilizations. However, there is little understanding of these palate varieties. In this paper, we present a large-scale study of recipes published on the web and their content, aiming to understand cuisines and culinary habits around the world. Using a database of more than 157K recipes from over 200 different cuisines, we analyze ingredients, flavors, and nutritional values which distinguish dishes from different regions, and use this knowledge to assess the predictability of recipes from different cuisines. We then use country health statistics to understand the relation between these factors and health indicators of different nations, such as obesity, diabetes, migration, and health expenditure. Our results confirm the strong effects of geographical and cultural similarities on recipes, health indicators, and culinary preferences across the globe.