https://Intuitive.zoom.us/j/96079672856?pwd=voqmtJqqGpz3BXC1UlYBnX05P475gu.1
 

Electrical Engineering Systems Seminar

Speaker: Moshe Yelisevitch

M.Sc. student under the supervision of Prof. Haim Yelisevitch

Wednesday, 25^th June 2025, at 15:00

Conditional Inverse Sampling for the Design of Antennas in Complex Environments

Abstract

The design of compact, high-performance antennas remains a formidable challenge due to the intricate relationship between structural geometry, material, and electromagnetic behavior. Traditional design approaches rely on iterative tuning and brute-force search, often requiring extensive electromagnetic (EM) simulations that are computationally expensive and time-consuming. Furthermore, real-world constraints such as environmental interactions, fabrication limitations, and nonlinear geometry-performance dependencies make it difficult to generate antennas that are both optimal and physically realizable. To address these challenges, we propose a novel Conditional Neural Inverse Transform Sampler (C-NITS) framework for inverse antenna design. Unlike conventional optimization-based approaches, our method learns to map desired electromagnetic characteristics, including reflection coefficient and radiation pattern, to a distribution of feasible antenna geometries, directly generating solutions that satisfy both performance and manufacturability constraints. Our approach extends the Neural Inverse Transform Sampler (NITS) to a conditional formulation, enabling controllable sampling based on environmental parameters such as substrate properties and nearby obstructions. By leveraging a learned inverse model and a fast surrogate simulation network, our framework efficiently explores high-dimensional design spaces without exhaustive full-wave EM simulations. A key feature of our method is its ability to generate diverse, multiple antenna solutions, rather than converging to a single optimal design. This enables engineers to explore multiple feasible configurations that satisfy design objectives while considering fabrication and environmental constraints. Furthermore, we incorporate a structured selection mechanism that filters generated designs to ensure real-world feasibility criteria. Our experiments demonstrate that C-NITS significantly outperforms traditional optimization techniques and existing deep-learning-based inverse design models in terms of EM similarity - both in numerical accuracy (e.g., pixel-wise comparisons) and structural similarity - as well as in engineering-relevant performance metrics. By combining conditional generative modeling with physics-aware constraints, our framework advances the state of automated antenna design, making it more adaptable and scalable for real-world applications.

השתתפות בסמינר תיתן קרדיט שמיעה = עפ"י רישום בצ'ט של שם מלא + מספר ת.ז.

https://tau-ac-il.zoom.us/j/82634921282?pwd=7Im8wd0FCJ7D0pIAMVoNglffXBaUrY.1
Meeting ID: 826 3492 1282
Passcode: 115437

Electrical Engineering Systems ZOOM Seminar

Speaker: David Uliel

M.Sc. student under the supervision of Prof. Raja Giryes

Sunday, 22^nd June 2025, at 15:00

Task Nuisance Filtration for Unsupervised Domain Adaptation

Abstract

In unsupervised domain adaptation (UDA) labeled data is available for one domain (Source

Domain) which is generated according to some distribution, and unlabeled data is available for a second domain (Target Domain) which is generated from a possibly different distribution but has the same task.

The goal is to learn a model that performs well on the target domain although labels are available only for the source data. Many recent works attempt to align the source and the target domains by matching their marginal distributions in a learned feature space. In this paper, we address the domain difference as a nuisance, and enables better adaptability of the domains, by encouraging minimality of the target domain representation, disentanglement of the features, and a smoother feature space that cluster better the target data. To this end, we use the information bottleneck theory and a classical technique from the blind source separation framework, namely, ICA (independent components analysis). We show that these concepts can

improve performance of leading domain adaptation methods on various domain adaptation benchmarks.

השתתפות בסמינר תיתן קרדיט שמיעה = עפ"י רישום בצ'ט של שם מלא + מספר ת.ז.

https://tau-ac-il.zoom.us/j/85063969240?pwd=t2xMC9bTXyQ2mtCTfTkywVCGJ8SvFc.1
Meeting ID: 850 6396 9240
Passcode: 529492

Electrical Engineering Systems Seminar

Speaker: Sol Yarkoni

M.Sc. student under the supervision of Prof. Roi Livni

Wednesday, 18^th June 2025, at 15:00

Low-Resource Reconstruction of Template-Memorized Images — Attack against Diffusion Models
Abstract

Diffusion models trained on large-scale datasets such as LAION have become foundational to modern generative AI. However, their reliance on uncurated web data introduces serious privacy risks, particularly through the unintended extraction of training images. This seminar presents the findings of the research paper “Low Resource Reconstruction Attacks Through Benign Prompts”, recently submitted to NeurIPS and an ICML workshop.

The talk introduces a low-resource image reconstruction attack capable of extracting template-memorized content from diffusion models using only black-box access and natural English prompts such as “Abstract Art T-Shirt.” The attack successfully reconstructs training images, including some containing real, identifiable individuals.

To contextualize the attack, the talk will also review the phenomenon of template memorization and examine how data scraped from e-commerce websites contributes to the image–text coupling hypothesized to underlie this behavior.

Sol Yarkoni is an M.Sc. student, experienced data scientist in computer vision, and generative artist.

השתתפות בסמינר תיתן קרדיט שמיעה = עפ"י רישום בצ'ט של שם מלא + מספר ת.ז.

הרישום לסמינר יבוצע באמצעות סריקת הברקוד למודל (יש להיכנס לפני כן למודל,  לא באמצעות האפליקציה)- הרישום מתחיל ב- 15:30 ומסתיים ב- 15:40

Registration to the seminar will be done by scanning the barcode for the Moodle (Please enter ahead to the Moodle, NOT by application)- Registration ends at 15:10

Electrical Engineering Systems Seminar

Speaker: Daniel Toma

M.Sc. student under the supervision of Dr. Wasim Huleihel

Sunday, 15th June 2025, at 15:30

Room 011, Kitot Building, Faculty of Engineering

Sequential Classification of Misinformation

In recent years there have been a growing interest in online auditing of information flow over social networks with the goal of monitoring undesirable effects, such as, misinformation and fake news. Most previous work on the subject, focus on the binary classification problem of classifying information as fake or genuine. Nonetheless, in many practical scenarios, the multi-class/label setting is of particular importance. For example, it could be the case that a social media platform may want to distinguish between ``true", ``partly-true", and ``false" information. Accordingly, in this thesis, we consider the problem of online multiclass classification of information flow. To that end, driven by empirical studies on information flow over real-world social media networks, we propose a probabilistic information flow model over graphs. Then, the learning task is to detect the label of the information flow, with the goal of minimizing a combination of the classification error and the detection time. For this problem, we propose two detection algorithms; the first is based on the well-known multiple sequential probability ratio test, while the second is a novel graph neural network based sequential decision algorithm. For both algorithms, we prove several strong statistical guarantees. We also construct a data driven algorithm for learning the proposed probabilistic model. Finally, we test our algorithms over two real-world datasets, and show that they outperform other state-of-the-art misinformation detection algorithms, in terms of detection time and classification error.