Integrated Calendar

Physiology or Medicine
Victor Ambros and Gary Ruvkun discovered microRNA, a new class of tiny RNA molecules that play a crucial role in gene regulation. Their groundbreaking discovery in the small worm C. elegans revealed a completely new principle of gene regulation. This turned out to be essential for multicellular organisms, including humans. MicroRNAs are proving to be fundamentally important for how organisms develop and function.

Physics
John Hopfield introduced a spin model that can store and reconstruct information. Geoffrey Hinton built on Hopfield’s idea to invent the Boltzmann Machine, that is able to learn from examples to reconstruct a set of desired patterns. He also popularized and improved Backpropagation of Errors, a method actually used in today’s advanced AI technology (e.g. Deep Learning).

Chemistry
The Nobel Prize in Chemistry 2024 is about proteins, life’s ingenious chemical tools. David Baker has succeeded with the almost impossible feat of building entirely new kinds of proteins. Demis Hassabis and John Jumper have developed an AI model to solve a 50-year-old problem: predicting proteins’ complex structures. These discoveries hold enormous potential.

Do we live in a simulation? Perhaps we should consider the possibility. Replicating real-world observations into a digital twin offers numerous potential benefits. For instance, in autonomous navigation, one could recreate safety-critical scenarios to test an agent's behavior more efficiently and without risking human lives. True-to-life simulations can enable counterfactual analysis, aiding in the interpretability of AI decision-making. Furthermore, they allow us to relive captured experiences for immersive entertainment.

In my research, I develop tools that enable these capabilities through the reconstruction of scene properties, such as geometry and color, and through perception methods like 3D scene segmentation and 3D object detection. This also includes the controllable generation and manipulation of content. To ensure scalability, my focus is particularly on representations that respect the symmetries in the data, such as rotation equivariance, and that can leverage large datasets at scale by minimizing the need for supervision.

Among these topics, in this talk, I will specifically highlight several of my recent papers. These include studies on 3D object detection from single images [1], neural fields for dynamic outdoor scene reconstruction [2], and piecewise equivariant representations [3].

[1] 3DiffTection: 3D Object Detection with Geometry-Aware Diffusion Features. Chenfeng Xu, Huan Ling, Sanja Fidler, Or Litany. CVPR 2024

[2] Zero-to-Hero: Enhancing Zero-Shot Novel View Synthesis via Attention Map Filtering, Ido Sobol, Chenfeng Xu, Or Litany. NeurIPS 2024

[3] EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision. Jiawei Yang, Boris Ivanovic, Or Litany, Xinshuo Weng, Seung Wook Kim, Boyi Li, Tong Che, Danfei Xu, Sanja Fidler, Marco Pavone, Yue Wang. ICLR 2024

Bio: Or Litany is a Senior Research Scientist at the NVIDIA Toronto AI research team, and an Assistant Professor at the Technion where he leads the LIT-Lab specializing in 3D computer vision and generative AI. He is an Azrieli Faculty Fellow and a Taub Fellow. Previously, he conducted postdoctoral research at Stanford University under the guidance of Prof. Leonidas Guibas, and at Meta AI Research (FAIR), where he was hosted by Prof. Jitendra Malik.

Micromotors/robots extend the reach of robotic operations to submillimeter dimensions and are becoming increasingly powerful for various tasks, such as the manipulation of micro/nanoscale cargo and single-cell analysis. These microrobots have the potential to significantly advance diagnostic testing and sample analysis, offering the benefits of traditional lab-on-a-chip devices (e.g., portability, efficiency) while overcoming current challenges (e.g., complexity, predetermined design, fluid control). Our recent findings have highlighted the unique advantage of using an electric field to enable unified, label-free, and selective micromotor-based cargo manipulation and transport [1]. Additionally, we have demonstrated the capability of electrically powered micromotors to (a) carry organelles or cells, (b) electro-deform cells as a novel means of biomechanical testing, and (c) electroporate cells for the transfection of drugs/genes [2]. Recently, the addition of magnetic field actuation has been shown to enable the operation of such hybrid-powered microrobots under near-physiological media conditions required for single-cell analysis [3]. Furthermore, optoelectronic control has been shown [4] to enable trajectory reconfiguration, directed self-assembly, and the parallelized operation of many such microrobots.
[1]Y. Wu, A. Fu & G. Yossifon, Small 1906682, 1-12 (2020).
[2]Y. Wu, A. Fu & G. Yossifon, PNAS 118, 38, e2106353118 (2021).
[3]Y. Wu, S. Yakov, A. Fu & G. Yossifon, Advanced Science 2204931 (2022).
[4]S. S. Das & G. Yossifon, Advanced Science 10, 2206183 (2023).

Capillary networks are prevalent in nature and biology, playing a crucial role in systems like animal vasculature and plant capillaries, with broad applications in medicine and science. However, many aspects of how these networks regulate and control flow remain unresolved. While the basic principles of capillary networks and their functions are well understood, ongoing research seeks to uncover how these systems dynamically respond to environmental changes, adapt to varying conditions, and whether they retain a memory of past states. Developing a model system for capillary networks allows us to pose exciting new questions, such as: "Can capillary networks store memory?"Building such a model presents two key challenges. First, the need to dynamically modify the nature of bonds within the networks and understand its impact on transport. Second, designing networks capable of evolving in response to external stimuli. Successfully addressing these challenges could transform our ability to actively control macroscale flow by manipulating local bonds within the networks.Here, a novel experimental model of capillary networks is proposed, consisting of hundreds of interconnected liquid diodes. Like electrical diodes, these microscale surface structures direct liquid flow in specific directions while preventing reverse flow. However, under certain conditions, liquid diodes may fail, permitting bidirectional flow and introducing bonds of varying properties within the capillary network.This system will allow us to investigate whether the wetting state of liquids in the network depends on its actuation history—essentially exploring whether capillary networks can exhibit memory. This question opens up new possibilities, including the potential to encode information within these networks, analyze how transport responds to external stimuli, study the interplay between global actuation and local fluid dynamics, explore the coupling between mechanics and flow, and better understand how information propagates through capillary systems.

Some of the most important open problems in group theory concern whether all groups can be metrically approximated by certain classes of groups. In particular, it is currently unknown whether there exist groups that are not (linear) sofic, hyperlinear, or MF.

 

We introduce the general problem of approximation of groups, as well as the related problem of stability, which asks whether almost homomorphisms from a group are close to actual homomorphisms. We recall a cohomological tool for stability introduced by De-Chiffre, Glebsky, Lubotzky, and Thom, which was used to prove the existence of groups that are not Frobenius-norm approximated. Moreover, we propose a weaker notion of semi-stability, which relates the different classes of approximated groups, and discuss how this cohomological tool can be adapted for this purpose.

 

For more details, check out our website.

We look forward to seeing you in one or both sessions next Monday.

The questions “How did life on Earth begin?” and “Are we alone in the universe?” are arguably two of the most intriguing in science. While until recently these questions tended to be relegated to the “too difficult” box, the attempts to answer them have now become extraordinarily vibrant and dynamic frontiers of science. I will describe how the quest for cosmic life follows two parallel, independent lines of research: cutting-edge laboratory studies aimed at determining whether life can emerge from pure chemistry, and advanced astronomical observations searching for signs of life on other planets and moons in the solar system and around stars other than the Sun. I will examine how using knowledge acquired through ingenious chemical experimentation, geological studies, advanced astronomical observations, and imaginative theorizing researchers have managed to delineate a plausible pathway leading from the formation of the Earth to the appearance of the early biological cells.