Pages

February 01, 2019

  • Date:30ThursdayJanuary 2025

    Vision and AI

    More information
    Time
    12:15 - 13:15
    Title
    Understanding and Enhancing Deep Neural Networks with Automated Interpretability
    Location
    Jacob Ziskind Building
    Room 1 - 1 חדר
    LecturerTamar Rott Shaham
    MIT
    Organizer
    Department of Computer Science and Applied Mathematics
    Contact
    AbstractShow full text abstract about Deep neural networks are becoming incredibly sophisticated; ...»
    Deep neural networks are becoming incredibly sophisticated; they can generate realistic images, engage in complex dialogues, analyze intricate data, and execute tasks that appear almost human-like. But how do such models achieve these abilities?

    In this talk, I will present a line of work that aims to explain behaviors of deep neural networks. This includes a new approach for evaluating cross-domain knowledge encoded in generative models, tools for uncovering core mechanisms in large language models, and their behavior under fine-tuning. I will show how to automate and scale the scientific process of interpreting neural networks with the Automated Interpretability Agent, a system that autonomously designs experiments on models’ internal representations to explain their behaviors. I will demonstrate how such understanding enables mitigating biases and enhancing models’ performance. The talk will conclude with a discussion of future directions, including developing universal interpretability tools and extending interpretability methods to automate scientific discovery.

    Bio: 

    Tamar Rott Shaham is a postdoctoral researcher at MIT CSAIL in Prof. Antonio Torralba’s lab. She earned her PhD from the ECE faculty at the Technion, supervised by Prof. Tomer Michaeli. Tamar has received several awards, including the ICCV 2019 Best Paper Award (Marr Prize), the Google WTM Scholarship, the Adobe Research Fellowship, the Rothchild Postdoctoral Fellowship, the Vatat-Zuckerman Postdoctoral Scholarship, and the Schmidt Postdoctoral Award.
    Lecture
  • Date:30ThursdayJanuary 2025

    Genomic and functional approaches to modeling and targeting cancer aneuploidy announced

    More information
    Time
    15:00 - 16:00
    Location
    Nella and Leon Benoziyo Building for Biological Sciences
    LecturerProf. Uri Ben-David, Prof. Uri Ben-David
    Dept of Human Molecular Genetics & Biochemistry, Faculty of Medicine, Tel-Aviv University
    Organizer
    Department of Biomolecular Sciences
    Contact
    AbstractShow full text abstract about Aneuploidy, an imbalanced number of chromosomes or chromosom...»
    Aneuploidy, an imbalanced number of chromosomes or chromosome arms, is a genetic hallmark of cancer cells, yet aneuploidy remains a biological enigma and a missed opportunity for cancer therapy. My lab applies experimental and computational approaches to dissect the basic biology underlying cancer aneuploidy and to study its cellular consequences. In this seminar, I will focus on an unpublished study in which we discovered an important role for a recurrent aneuploidy in driving brain metastasis, revealed the underlying molecular mechanism, and identified a therapeutically-relevant cellular vulnerability that is associated with this common aneuploidy.
    Lecture
  • Date:30ThursdayJanuary 2025

    Special Clore Seminar - The Clore Center for Biological Physics

    More information
    Time
    16:15 - 18:00
    Title
    The Physics of Ancient Chromosomes
    Location
    Nella and Leon Benoziyo Physics Library
    LecturerProf. Erez Lieberman Aiden
    Refreshments at 16:00
    Organizer
    Clore Center for Biological Physics
    Contact
    AbstractShow full text abstract about We report that skin from a female woolly mammoth (†Mammuthus...»
    We report that skin from a female woolly mammoth (†Mammuthus primigenius) that died 52, 000 years ago retained its ancient genome architecture. We hypothesize that, shortly after this mammoth’s death, the sample spontaneously freeze-dried in the Siberian cold, leading to a glass transition that preserved subfossils of ancient chromosomes at nanometer scale. This talk will explore the physics that underlies the preservation of ancient chromosomes. FOR THE LATEST UPDATES AND CONTENT ON SOFT MATTER AND BIOLOGICAL PHYSICS AT THE WEIZMANN, VISIT OUR WEBSITE: https://www.biosoftweizmann.com/
    Lecture
  • Date:02SundayFebruary 2025

    Methane mitigation by unique redox couplings in freshwater sediments

    More information
    Time
    11:00 - 12:00
    Location
    Sussman Family Building for Environmental Sciences
    M. Magaritz seminar room
    LecturerOrit Sivan
    Organizer
    Department of Earth and Planetary Sciences
    AbstractShow full text abstract about This talk tests the ability of natural freshwater lakes and ...»
    This talk tests the ability of natural freshwater lakes and margins to attenuate the emissions ofthe greenhouse gas methane (CH4) to the atmosphere under warming climate. I will show howmicrobial communities manage to survive and mitigate methane emissions under energylimited, highly reduced conditions of deep methanogenic lake sediments, through redoxcouplings of methane to Mn-Fe-N. Complex redox couplings between those species were alsoexplored in thermokarst lakes and margins, which are extensively formed by permafrost thawin the Arctic. The cycles were quantified using geochemical and microbial profiles, togetherwith stable isotope probing experiments close to natural conditions. The profiles andincubations show active microbial population that exhibit surprisingly both aerobic andanaerobic methane oxidation in methanogenic sediments and upland Arctic soils, fueled bynitrogen and iron redox cycles.
    Lecture
  • Date:02SundayFebruary 2025

    The Clore Center for Biological Physics

    More information
    Time
    12:45 - 14:30
    Title
    Design principles of protein-DNA Recognition Specificity in Embryonic Stem Cells
    Location
    Nella and Leon Benoziyo Physics Building
    LecturerDavid B. Lukatsky
    Lunch at 12:45
    Organizer
    Clore Center for Biological Physics
    Contact
    AbstractShow full text abstract about Transcription factors (TFs) bind genomic DNA regulating gene...»
    Transcription factors (TFs) bind genomic DNA regulating gene expression and developmental programs in embryonic stem cells (ESCs). Even though comprehensive genome-wide molecular maps for TF-DNA binding are experimentally available for key pluripotency-associated TFs, the understanding of molecular design principles responsible for TF-DNA recognition remains incomplete. In this talk, I will show that binding preferences of key pluripotency TFs exhibit bimodality in the local GC-content distribution. Sequence-dependent binding specificity of these TFs is distributed across three major contributions. First, local GCcontent is dominant in high-GC-content regions. Second, recognition of specific k-mers is predominant in low-GC-content regions. Third, short tandem repeats (STRs) are highly predictive in both low- and high-GC-content regions. In sharp contrast, binding preferences of a key oncogenic protein, c-Myc, are exclusively dominated by local GC-content and STRs in high-GC-content genomic regions. I will propose that the transition in the TF-DNA binding landscape upon ESC differentiation is solely regulated by the concentration of c-Myc, which forms a bivalent c-Myc-Max heterotetramer upon promoter binding, competing with key pluripotency factors. Taken together, these findings point out that c-Myc may significantly affect the genome-wide TF-DNA binding landscape, chromatin structure, and enhancerpromoter interactions.FOR THE LATEST UPDATES AND CONTENT ON SOFT MATTER AND BIOLOGICAL PHYSICS AT THE WEIZMANN, VISIT OUR WEBSITE: https://www.biosoftweizmann.com/
    Lecture
  • Date:03MondayFebruary 2025

    Weizmann neuroscience research symposium

    More information
    Time
    08:00 - 08:00
    Location
    The David Lopatie Conference Centre
    Chairperson
    Rony Paz
    Organizer
    The Azrieli Institute for Brain and Neural Sciences
    Conference
  • Date:03MondayFebruary 2025

    New concepts, catalysts and methods in synthetic chemistry

    More information
    Time
    11:00 - 12:15
    Location
    Gerhard M.J. Schmidt Lecture Hall
    LecturerProf. Mark Gandelman
    AbstractShow full text abstract about In my lecture, I will showcase how designing new materials a...»
    In my lecture, I will showcase how designing new materials and exploring their fundamental properties can lead to innovative concepts and practical applications in organic chemistry. We will begin by discussing the synthesis of novel halo-organic compounds that enable the stereoselective catalytic synthesis of biologically relevant chiral organofluorides.     The talk will primarily focus on the versatile chemistry of N-Heterocyclic Nitrenium ions (NHNs) – the nitrogen-based analogs of ubiquitous N-Heterocyclic Carbenes. We will demonstrate their unique coordination abilities, analyze their properties, and highlight their role in stabilizing elusive species.1,2 Nitrenium ions represent a novel family of nitrogen-based Lewis acids3 and serve as efficient metal-free catalysis, frustrated Lewis pairs partners4 and platform for isolating valuable radicals.5 Finally, we will demonstrate how the fundamental understanding nitrenium properties led to the development of triazenolysis reaction - an aza-version of the canonical alkene ozonolysis.6References:[1] Nat. Chem. 2011, 5, 525.[2] Chem.Sci. 2014, 5, 1305.[3] J. Am. Chem. Soc. 2017, 139, 4062.[4] Angew. Chem. Int. Ed. 2020, 59, 23476.[5] J. Am. Chem. Soc. 2022, 144, 23642; J. Am. Chem. Soc. 2024, 146, 19474.[6] Nat. Chem. 2025, 17, 101.
    Colloquia
  • Date:03MondayFebruary 2025

    Foundations of Computer Science Seminar

    More information
    Time
    11:15 - 12:15
    Title
    Vizing's Theorem in Near-Linear Time
    Location
    Jacob Ziskind Building
    Room 1 - 1 חדר
    LecturerShay Solomon
    TAU
    Organizer
    Department of Computer Science and Applied Mathematics
    Contact
    AbstractShow full text abstract about Vizing's Theorem from 1964 states that any n-vertex m-e...»
    Vizing's Theorem from 1964 states that any n-vertex m-edge graph of maximum degree Δ can be edge colored using at most Δ+1 different colors.

    Vizing's original proof is algorithmic and implies that such an edge coloring can be found in O(mn) time.

    In this talk, I'll present a randomized algorithm that computes a (Δ+1)-edge coloring in near-linear time -- in fact, only O(mlogΔ) time -- with high probability.
    Lecture
  • Date:04TuesdayFebruary 2025

    From chromosomes to single genes: Designing DNA molecules for autonomous cell-free systems

    More information
    Time
    14:00 - 15:00
    Location
    Gerhard M.J. Schmidt Lecture Hall
    LecturerDr. Ferdinand Greiss
    Organizer
    Department of Chemical and Structural Biology
    Lecture
  • Date:05WednesdayFebruary 2025

    2025 Israeli Nucleic Acids Therapeutics Meeting

    More information
    Time
    08:30 - 18:00
    Location
    The David Lopatie Conference Centre
    Chairperson
    Igor Ulitsky
    Organizer
    Abisch-Frenkel RNA Therapeutics Center
    Conference
  • Date:05WednesdayFebruary 2025

    Azrieli institute for brain and neural sciences students seminar

    More information
    Time
    10:30 - 12:00
    Title
    students seminar
    Location
    Arthur and Rochelle Belfer Building for Biomedical Research
    Botnar auditorium
    Organizer
    The Azrieli Institute for Brain and Neural Sciences
    Lecture
  • Date:05WednesdayFebruary 2025

    Life Sciences Luncheon

    More information
    Time
    12:00 - 14:00
    Title
    Prof. Tamar Geiger
    Location
    Nella and Leon Benoziyo Building for Biological Sciences
    Auditorium
    LecturerProf. Tamar Geiger
    Contact
    Lecture
  • Date:06ThursdayFebruary 2025

    Novel informatics approaches to solving biological puzzles

    More information
    Time
    09:00 - 10:00
    Location
    Candiotty
    Auditorium
    LecturerDr. Danny Ben-Avraham, Dr. Avital Sarusi-Portugues, Dr. Barak Markus, Dr. Efrat Ben-Zeev
    Bioinformatics unit, G-INCPM
    Lecture
  • Date:06ThursdayFebruary 2025

    Vision and AI

    More information
    Time
    12:15 - 13:15
    Title
    Leveraging Pretrained Generative Models for Real Image Editing
    Location
    Jacob Ziskind Building
    Room 1 - 1 חדר
    LecturerOr Patashnik
    Tel Aviv University
    Organizer
    Department of Computer Science and Applied Mathematics
    Contact
    AbstractShow full text abstract about Image generative models are advancing rapidly, producing ima...»
    Image generative models are advancing rapidly, producing images of remarkable realism and fidelity. However, existing models often lack precise control over the generated content, limiting their image editing capabilities and the integration of real content into synthesized imagery. In this talk, I will demonstrate how a deep understanding of the inner mechanisms of large-scale pretrained generative models enables the design of powerful techniques for a variety of image manipulation tasks. By analyzing the semantic representations learned by these models, I will present methods that enable effective content editing. Additionally, I will discuss the challenges and trade-offs involved in manipulating real content and propose strategies to address these challenges. Finally, I will highlight recent advancements in incorporating real content, with a particular focus on techniques for injecting information into pretrained models.

    Bio: Or Patashnik (https://orpatashnik.github.io/) is a Computer Science PhD candidate at Tel Aviv University, supervised by Daniel Cohen-Or. Her research focuses on computer graphics and its intersection with computer vision, with an emphasis on generative tasks such as image editing, personalization, and image inversion using large-scale pretrained models. Recently, she has been particularly interested in better understanding diffusion models for various applications.
    Lecture
  • Date:06ThursdayFebruary 2025

    Senescent cells on the crossroads of cancer and aging

    More information
    Time
    14:00 - 15:00
    Location
    Candiotty Auditorium
    LecturerProf. Valery Krizhanovsky
    Organizer
    Dwek Institute for Cancer Therapy Research , Moross Integrated Cancer Center (MICC)
    Lecture
  • Date:09SundayFebruary 2025

    Climate and Solar Variability: A Critical Evaluation

    More information
    Time
    11:00 - 12:00
    Location
    Sussman Family Building for Environmental Sciences
    M. Magaritz seminar room
    LecturerNathan Steiger
    Contact
    AbstractShow full text abstract about It has been claimed that solar variability is the largest dr...»
    It has been claimed that solar variability is the largest driver of climatic variability across thePhanerozoic eon and that it accounts for ½ to ⅔ of20th century warming. Apparent evidence insupport of these claims has been mustered frommodeling work along with paleoclimatic, oceanographic, and other observational datasets.Here I will show that this research fails to replicate. I additionally find that many studies claiming tosupport a strong solar-climate link suffer from fundamental statistical and conceptual errors thatinvalidate their results.
    Lecture
  • Date:09SundayFebruary 2025

    The Clore Center for Biological Physics

    More information
    Time
    12:45 - 14:15
    Title
    The role of promiscuous molecular recognition in the evolution of self-incompatibility in plants
    Location
    Nella and Leon Benoziyo Physics Library
    LecturerDr. Tamar Friedlander
    Lunch at 12:45
    Organizer
    Clore Center for Biological Physics
    Contact
    AbstractShow full text abstract about How do biological networks evolve and expand? We study these...»
    How do biological networks evolve and expand? We study these questions in the context of the plant collaborative-non-self recognition self-incompatibility system. Self-incompatibility evolved to avoid self-fertilization among plants. It relies on specific molecular recognition between highly diverse proteins expressed in the female and male reproductive organs, such that the combination of proteins an individual possesses determines its mating partners, defining distinct ‘mating specificities’. Although a few dozen mating specificities are known from population surveys, previous models struggled to pinpoint the evolutionary trajectories by which new specificities evolved. We construct a novel theoretical framework, synthesizing evolutionary and biophysical models, that crucially affords interaction promiscuity and multiple distinct partners per protein, as is seen in empirical findings. We demonstrate spontaneous self-organization of the population into distinct 'classes' with full between-class compatibility and a dynamic long-term balance between class emergence and decay.Our work highlights the importance of molecular recognition promiscuity to network evolvability. Promiscuity was found in additional systems suggesting that our framework could be more broadly applicable. FOR THE LATEST UPDATES AND CONTENT ON SOFT MATTER AND BIOLOGICAL PHYSICS AT THE WEIZMANN, VISIT OUR WEBSITE: https://www.biosoftweizmann.com/  
    Lecture
  • Date:09SundayFebruary 2025

    PhD Thesis Defense - Noam Galili Darnell -Prof Itay Tirosh lab

    More information
    Time
    13:00 - 15:00
    Title
    Deciphering Glioma Organization and Chromosome Copy Number Variability
    Location
    Zoom
    Lecture
  • Date:11TuesdayFebruary 2025

    Tu Bishvat event

    More information
    Time
    10:00 - 12:00
    Location
    Michael Sela Auditorium
    Cultural Events
  • Date:11TuesdayFebruary 2025

    Special Guest Seminar

    More information
    Time
    11:00 - 12:00
    Location
    Max and Lillian Candiotty Building
    Auditorium
    LecturerDr. Omri Wurtzel
    Organizer
    Department of Immunology and Regenerative Biology
    Contact
    Lecture

Pages