הפקולטה להנדסה ע"ש איבי ואלדר פליישמן

EE Seminar: Two Asymmetric Descriptions from Many Symmetric Descriptions

Speaker: Adam Mashiach,

M.Sc. student under the supervision of Prof. Ram Zamir.

Wednesday, December 7^th, 2016 at 15:00

Room 011, Kitot Bldg., Faculty of Engineering

Two Asymmetric Descriptions from Many Symmetric Descriptions

Abstract

Multiple descriptions provide a mechanism for graceful degradation in open-loop lossy source coding over a lossy-packet network. Existing solutions for the general asymmetric case are, however, cumbersome. We present a framework for designing an efficient asymmetric two-description scheme, based on dividing many symmetric descriptions into two groups, and encoding each group jointly.

In the first part, we present a multiple description coding scheme which is based on oversampling and dithered delta-sigma quantization that can generate large number of symmetric descriptions. In the second part, we use this symmetric many description scheme to build an asymmetric sum-rate optimal two-description scheme. We further show that this idea can be extended to generate any number asymmetric descriptions (not only two), which are sum-rate optimal for an interesting specific case of the many description problem.

07 בדצמבר 2016, 15:00

חדר 011, בניין כיתות-חשמל

הרצאת אורח בסמינר המחלקה להנדסה ביו רפואית

Prof. Elisha Moses

Physics of Complex Systems, Weizmann Institute of Science, Israel

11 בדצמבר 2016, 15:00

Magnetic Stimulation of Neural Networks and the Brain

Prof. Elisha Moses

Physics of Complex Systems, Weizmann Institute of Science, Israel

External stimulation of the brain is emerging as a novel methodology for treatment of mental illness and possibly also for cognitive enhancement. Electric and magnetic and even ultrasound stimulation of neurons have all shown to be effective in eliciting brain activation, but the actual effect on neurons remains very unclear. Combining experiments on excitation in neuronal cultures, animals and humans with theory and numerical simulations, we have been able to unravel the contribution of electric and of magnetic pulses delivered to the brain. We show that today’s magnetic stimulation techniques do not optimally target neurons in the brain, and that they can be considerably enhanced with simple technical modifications involving rotating magnetic fields and prolonged pulse durations. We end by suggesting practical clinical trials for the near future.

ההרצאה תתקיים ביום ראשון 11.12.16, בשעה 15:00

בחדר 315, הבניין הרב תחומי, אוניברסיטת תל אביב

תלמידי המחזור הראשון בתכנית המצטיינים "הנדסה-רוח" במפגש תחילת שנה.

קרא עוד אודות תלמידי המחזור הראשון בתכנית המצטיינים "הנדסה-רוח" במפגש תחילת שנה.

EE Seminar: Computational challenges in protein-RNA interactions

(The talk will be given in English)

Speaker: Dr. Yaron Orenstein
Massachusetts Institute of Technology

Monday, December 26^th, 2016
15:00 - 16:00

Room 011, Kitot Bldg., Faculty of Engineering

Computational challenges in protein-RNA interactions

Abstract

Protein-RNA binding, mediated through both RNA sequence and structure, plays vital role in many cell processes, including neurodegenerative-diseases. Modeling the sequence and structure binding preference of an RNA-binding protein is a key computational challenge. Accurate models will enable prediction of new interactions and better understanding of the binding mechanism. In addition, designing compact and efficient sequence libraries to experimentally measure these interactions is necessary to discover novel binding preferences. In this talk, I will present my work in solving these two challenges. In the first part, I will describe RCK, an efficient algorithm to learn k-mer based sequence and structure scores, which outperforms the state-of-the-art. I will give examples of novel biological insights we can gain by applying RCK to the largest dataset of protein-RNA interactions. In the second part, I will consider the problem of generating a minimum-size set of unstructured RNA sequences covering all k-mers. I will prove that a general definition of this problem is NP-hard, and describe CurlCAKE, a greedy heuristic to solve this problem that works well in practice. I will conclude with open questions and future plans.

Bio

Yaron’s research interests lie in Bioinformatics. Yaron’s main focus is on the development of efficient algorithms to infer accurate binding models from experimental data and algorithms to generate compact universal experimental designs.

Yaron is currently a post-doc at CSAIL, MIT, working with Prof. Bonnie Berger. Previously, he received his Ph.D from CS, TAU, where he worked under the supervision of Prof. Ron Shamir. He received his M.Sc degree from EE, TAU, where he worked under the supervision of Prof. Dana Ron. Before that, Yaron completed his bachelor’s degree in computer science and electrical engineering at TAU.

26 בדצמבר 2016, 15:00

חדר 011, בניין כיתות-חשמל

EE Seminar: Nonparametric Canonical Correlation Analysis

(The talk will be given in English)

Speaker: Dr. Tomer Michaeli
EE, Technion, Haifa

Monday, December 5^th, 2016
15:00 - 16:00

Room 011, Kitot Bldg., Faculty of Engineering

Nonparametric Canonical Correlation Analysis

Abstract

Canonical correlation analysis (CCA) is a classical representation learning technique for finding correlated variables in multi-view data. This tool has found widespread use in various fields, including recent application to natural language processing, speech recognition, genomics, and cross-modal retrieval. One of the shortcomings of CCA is its restriction to linear mappings, since many real-world multi-view datasets exhibit highly nonlinear relationships. In recent years, several nonlinear extensions of the original linear CCA have been proposed, including kernel and deep neural network methods. These approaches significantly improve upon linear CCA in many practical applications, but have three limitations. First, they are still restricted to families of projection functions which the user must specify (by choosing a kernel or neural network structure). Second, they are computationally demanding. Third, they often produce redundant representations.
In this work, we derive a closed form solution to the nonlinear CCA problem without any functional restrictions. As we show, the solution corresponds to the SVD of a certain operator associated with the joint density of the views. Thus, by estimating the population density from training data, we obtain a practical nonparametric CCA (NCCA) algorithm, which reduces to solving an eigenvalue system. Superficially, this is similar to kernel CCA, but importantly, NCCA does not require the inversion of any kernel matrix. We also derive a partially linear CCA (PLCCA) variant in which one of the views undergoes a linear projection while the other is nonparametric. Finally, we show why spectral representation learning methods may produce redundant representations, and propose a generic technique to prevent redundancy in those algorithms.
As we demonstrate on several test cases, our algorithms are memory-efficient, often run much faster and perform better than kernel CCA and comparable to deep CCA.

This is joint work with Weiran Wang, Karen Livescu and Yochai Blau.

05 בדצמבר 2016, 15:00

חדר 011, בניין כיתות-חשמל

יום המחלקה להנדסה ביו רפואית

SAVE THE DATE 25.12.16

25 בדצמבר 2016, 13:00

יום מחלקה במרכז לארץ ישראל יפה

חובה להירשם במייל anatba@tauex.tau.ac.il

קול קורא לבוגרי הפקולטה CIM

קול קורא לבוגרי הפקולטה - בואו לשמוע ולהשפיע

מעבדת CIM - computer integrated manufacturing

מארחת את בוגרי הפקולטה

יום רביעי 04.01.17 בין השעות 18:30-20:30

המפגש מיועד לבוגרי הפקולטה להנדסה, אוניברסיטת תל אביב בלבד

נדרשת הרשמה מראש

לינק לפרטים והרשמה

במפגש, הצגת המעבדה והנעשה בה עם צוות המעבדה ודוקטורנטים.

כולל דיון ושיח פתוח על מנת להיעזר בניסיון שלכם. לשמוע מהם הנושאים שאתם הייתם ממליצים ללמד במעבדה

04 בינואר 2017, 18:30

בניין וולפסון בפקולטה להנדסה אוניברסיטת תל אביב

בתכנית

הצגת המעבדה
הצגת סרטי פרוייקטים של סטודנטים
סקירה קצרה של מצב עכשווי
דיון וסיעור מוחות להצעת נושאי לימוד נוספים ולעדכון נושאים קיימים

כמה מילים על המעבדה

המעבדה עוסקת במגוון נושאים הקשורים בתעשייה המודרנית, כשהמכנה המשותף שלהם הוא טכנולוגיה, בקרה ומחשוב. מרבית נושאים אלו ניתן לכלול תחת הכותרת של "מערך ייצור או שירות חכם" והם כוללים מרכיבים כגון רובוטיקה, מערכות הרכבה אוטומטית, תיכון תהליכים ממוחשב, מערכות ייצור אינטגרטיבי, בקרה ממוחשבת, תכנות, חיישנים, רובוטיקה, סימולציות ועוד.

הרצאת אורח המחלקה להנדסה ביו רפואית

27 בנובמבר 2016, 15:00

Sperm Morphological Analysis with Interferometric Phase Microscopy

Dr. Michal Balberg

Visiting Scholar, Dept. of Biomedical Engineering, Tel Aviv University

Senior Lecturer, Holon Institute of Technology, Israel

Sperm cell morphology is analyzed for assessing the fertility potential of a male. Currently, it is performed by staining the cells with a dye that creates a contrast between morphological features, such as the nucleus and the acrosome, and imaging it with a bright field microscope. Interferometric phase microscopy (IPM) is an imaging technique capable of capturing both amplitude and phase maps of transparent biological cells, and extracting the optical thickness of the cell. We show that according to the WHO criteria for morphological analysis, IPM provides equivalent information to the current practice. Following, we show that we can extract, without staining, other physiological parameters, such as the dry mass of the cell. These initial findings may assist in selecting unstained sperm cells for in-vitro cytoplasmic injection (ICSI) in the future, based on their morphological features, where staining is not allowed.

ההרצאה תתקיים ביום ראשון 27.11.16, בשעה 15:00

בחדר 315, הבניין הרב תחומי, אוניברסיטת תל אביב

סמינר המחלקה להנדהס ביו רפואית

27.11.16

27 בנובמבר 2016, 14:30

עידו וינר

תלמיד התואר השני במחלקה לביולוגיה מולקולרית ואקולוגיה של צמחים

אוניברסיטת תל אביב

Towards Microalgae-based Technologies
Harnessing Computational Models to Overcome the Heterologous Gene Expression Barriers in Microalgae

Recently, various species of microalgae have emerged as promising host-organisms for biotechnology industries, due to a set of unique features. These include rapid cell division rates, efficient conversion of sunlight to organic compounds by photosynthesis, and the ability to grow in extreme conditions. However, the inability to receive high levels of heterologous gene expression in microalgae is hindering the development of the entire field. Indeed, most current microalgae-based technologies focus on producing substances that occur naturally in the host species.

The chloroplast genome is considered a promising target for the insertion of foreign genes due to the lack of silencing mechanisms and the relatively high overall expression of chloroplast genes. Although some successful attempts to effectively produce proteins of interest in the chloroplast genome have been recorded, a better understanding of the mechanisms regulating translation is required. Current knowledge is mainly based on empirical mutation assays carried out for several genes in a small set of chloroplast genomes. Very few large-scale analyses of features and signals related to gene expression mechanisms in the chloroplast genome were reported, thus limiting the robustness and scope of the conclusions made so far.

To this end we have comprised a data-set of chloroplast genomes and applied computational models and motif-searching algorithms to detect novel global features on the one hand, and improve the current understanding of known motifs on the other hand. We provide new insights into the role of the bacterial Shine-Delgarno element in chloroplasts. Subsequently, we were able to detect additional motifs containing signals that suggest their involvement in new unknown canonical translation initiation mechanisms.

העבודה נעשתה בהנחיית ד"ר יפתח יעקובי המחלקה לביולוגיה מולקולרית ואקולוגיה של צמחים ופרופ' תמיר טולר המחלקה להנדסה ביו-רפואית אוניברסיטת תל-אביב

ההרצאה תתקיים ביום ראשון 27.11.16, בשעה 14:30

בחדר 315, הבניין הרב תחומי, אוניברסיטת תל אביב

EE Seminar: Diversity Multiplexing Tradeoff Under Generalized Sampling Expansion

Speaker: Nir Hadas

M.Sc. student under the supervision of Prof. Meir Feder

Wednesday, December 7^th, 2016 at 15:30

Room 011, Kitot Bldg., Faculty of Engineering

Diversity Multiplexing Tradeoff Under Generalized Sampling Expansion

Abstract

Standard analysis of Single Input Multiple Output (SIMO) and Multiple Input Multiple Outout (MIMO) channels with N output signals assumes sampling each of the output signals at the Nyquist rate. Yet it is known from the Generalized Sampling Expansion (GSE) that for a SIMO channel a unique reconstruction of the input signal is possible when each output signal is sampled at 1/N the Nyquist rate. Also, from the Vector Sampling Expansion (VSE), it is known that fora MIMO channel with M input signals and N received signals, a unique reconstruction of the input signals is possible when each output signal is sampled at M/N the Nyquist rate, under the condition that N/M is an integer.

It is a common practice to evaluate fading channels by their Diversity Multiplexing Tradeoff (DMT).

In this work we calculate and analyze the optimal DMT of the 1xN GSE-sampled SIMO fading channel an achievable DMT for the MxN VSE-sampled MIMO channel.

07 בדצמבר 2016, 15:30

חדר 011, בניין כיתות-חשמל

עמודים