У уторак, 27. марта у 13.30 часова у читаоници библиотеке „Др Драган Поповић“ Др Владан Вулетић са Департмана за физику МИТ-а одржаће предавање

„Laser cooling of Rb to Bose-Einstein condensation“

АПСТРАКТ:

The procedure for attaining quantum degeneracy in atomic gases almost exclusively rely on a final evaporative cooling, a time-consuming step associated with substantial atom loss. We demonstrate direct laser cooling of a gas of 87Rb atoms to quantum degeneracy. The method is fast and induces little atom loss. The atoms are trapped in a two-dimensional optical lattice that enables cycles of compression to increase the density followed by Raman sideband cooling to decrease the temperature. From a starting number of 2000 atoms, 1400 atoms reach quantum degeneracy in 300 milliseconds, as confirmed by a bimodal velocity distribution. The method should be broadly applicable to many bosonic and fermionic species and to systems where evaporative cooling is not possible.

У оквиру SCL семинара Центра за студије комплексних система, у понедељак, 26. марта у 14 часова у читаоници библиотеке „Др Драган Поповић“ предавање ће одржати др Александру Николин („Хорија Хулубеј“ национални институт за физику и нуклеарни инжињеринг, Букурешт, Румунија).

Др Николин ће одржати предавање: „How to test a voice biometrics solution: statistics of the Romanian and Greek languages“

АПСТРАКТ:

Motivated by the development of a novel voice-biometrics solution within the H2020 SpeechXRays project, we will present a series of statistical results on the Romanian and Greek languages that can be used to check the accuracy of voice biometrics solutions across the phonetic spectrum of the aforementioned languages. We will revisit the concept of phonetic balance and argue that in addition to the correct distribution of phonemes, a good set of test utterances should also have the correct distribution of clusters of phonemes and should account for familiarity and comfort of the users with the individual words within the test utterances. 

у понедељак, 19. марта 2018. у читаоници библиотеке „Др Драган Поповић“, у 13:30 часова Марко Цетина ће одржати предавање:

   An Error-corrected, Universal, Re-configurable, Ion-trap Quantum Archetype

Институт за физику позива вас на предавање проф. Кунрада Схалма са Лоренц института за теоријску физику Универзитета Лајден у Холандији. Предавање под насловом „(Quantum) Chaos, hydrodynamics, and black hole scrambling“ у оквиру традиционалног колоквијума Института за физику у Београду биће одржано у згради Института, у уторак, 20. марта у 11 часова у сали „Звонко Марић“.

АПСТРАКТ:

A concrete definition of chaos in quantum systems is a regime of exponential growth in a specific out-of-time-ordered correlator (OTOC). We will discuss this both in weakly coupled and in strongly coupled theories. In strongly coupled quantum theories with a holographic dual, the OTOC is directly related to probing the system’s hydrodynamic sound modes. The information contained in the OTOC is dual to a gravitational shock wave and this can be reconstructed in terms of purely diffusion-like gravitational waves at the horizon of a single-sided black hole with specific regularity-enforced imaginary values of frequency and momentum. The fact  that late-time equilibration and early-time chaos are controlled by the same dynamics is also a property of classical dilute gases. For weakly coupled systems, we show that the information contained in the OTOC follows from a Boltzmann kinetic equation for total energy exchange.

Центар за изучавање комплексних система организује SCL семинар у петак, 9. марта 2018. у 14 часова у читаоници библиотеле „Др Драган Поповић“. На семинару ће говорити др Јаћим Јаћимовић из ABB Corporate Research, Швајцарска.

Тема семинара је: „Net Shape 3D Printed NdFeB Permanent Magnet“

АПСТРАКТ:

Nd2Fe14B hard magnets are a strategic material in determining the efficiency and size of both existing and future electrical and electronics devices. At this talk, we will present firstly that heavy rare earth element Dy that improve temperature stability of magnet, but lowers magnetic properties, may not be required throughout the full magnet, and the steps in experimental realization of such magnet geometry will be discussed. Second, we will show that the emerging technology named 3D metal printing by selective laser melting can be used to realize hard magnets in various shapes, with novel functionalities. The magnetic properties of 3D printed will be compared to those prepared by other methods.

Центар за изучавање комплексних система организује SCL семинар у четвртак, 8. марта 2018. у 14 часова у читаоници библиотеле „Др Драган Поповић“. На семинару ће говорити Владимир Глигоријевић са Flatiron института у Њујорку, САД.

Тема семинара је: „Deep Multi-network Embedding for Protein Function Predictiction“.

АПСТРАКТ:

The prevalence of high-throughput experimental methods has resulted in an abundance of large-scale molecular and functional interaction networks.  The connectivity of these networks provides a rich source of information for inferring functional annotations for genes and proteins. An important challenge has been to develop methods for combining these heterogeneous networks to extract useful protein feature representations for function prediction. Most of the existing approaches for network integration use shallow models that cannot capture complex and highly-nonlinear network structures. We introducedeepNF, our novel deep-learning based network integration method for protein function prediction. deepNF consists of two steps: 1) creating a low-dimensional dense vector representation of proteins (i.e., embedding) using Multimodal Deep Autoencoders and 2) training a classifier on the resulting representation to predict protein functions. 

We apply deepNF on 6 different networks obtained from the STRING db to construct a compact low-dimensional representation containing high-level protein features. We will present an extensive performance analysis comparing our method with the state-of-the-art network integration methods for protein function prediction. In addition to cross-validation, the analysis also includes a temporal holdout validation evaluation similar to the measures in Critical Assessment of Functional Annotation (CAFA). Our results show that our method outperforms previous methods for both human and yeast STRING networks. Our method offers a great advantage of being able to capture non-linear information conveyed by large-scale biological networks, leading to improved network representations. Features learned by our method lead to substantial improvements in protein function prediction accuracy, which could enable novel protein function discoveries.