October 2021 - semfis - Mailing lists@ IFISC

IFISC Seminar on Wednesday, Nov 03, 2021 at 14:30
by semfis＠ifisc.uib-csic.es 29 Oct '21

29 Oct '21

IFISC Seminar on Wednesday Nov 03, 2021 at 14:30 Place: IFISC Seminar Room Title: Photonic reservoir computing in optical communications: does it have a chance against digital signal processing? Speaker: Apostolos Argyris, IFISC Abstract: The latest improvements in optical communications include always digital signal processing assistance. Machine learning and neural networks were also incorporated, to enforce problem-solving in tasks that are demanding in terms of complexity and computational speed. These algorithms aim at expanding the reach of fiber transmission channels, improving data recovery and optical routing. In this context, reservoir computing (RC) inspired hardware photonic implementations that can offer revolutionary solutions. The latter were initially tested in wireless channel equalization. Lately, they were endorsed for modulation format identification, optical header recognition, and data recovery. Here, I discuss the requirements that analog, photonic RC implementations need to fulfil for real-time signal processing in optical communicati ons. While offline processing was successfully demonstrated, the challenges to address the data rates of the latest communication protocols are still significant. <a href="https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09">https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09</a> Url event:https://ifisc.uib-csic.es/en/events/seminars/photonic-reservoir-computing-in-optical-communicat/

1 0

Reminder: Master Thesis, TODAY - Wednesday, Oct 27, 2021 at 09:30
by semfis＠ifisc.uib-csic.es 27 Oct '21

27 Oct '21

Master Thesis on Wednesday Oct 27, 2021 at 09:30 Place: Zoom Seminar Title: Evolutionary Game Theory with payoff fluctuations: Bet-Hedging Speaker: Rubén Calvo Ibáñez, IFISC/UIB Abstract: This thesis focuses on bet-hedging in changing evolutionary games. We study games with environmental fluctuations and quantify the effect that the distribu- tion of fitness of a given strategy has on its success. In general, it is an accepted idea in the field of biology, that reducing the variance of this distribution, even at the expense of reducing its mean, can be beneficial. We model this scenario as a simple game having stochastic fluctuations in the payoff matrix, and analyze it in detail. Our simulation results confirm the existence of bet-hedging strategies, in general, for a wide range of distributions. However, the particular asymmetry that the distribution has is seen to play an important role. We were able to demonstrate that there exist new types of bet-hedging when the distribution is negatively skewed. These are characterized by an increase in the variance of fitness fluctuations leading to higher success, even at the expense of having to reduce mean fitness. Jury: Pere Colet, Manuel Matias,&nbsp;Tobias Galla Advisor: Tobias Galla Zoom:&nbsp;https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09 Url event:https://ifisc.uib-csic.es/en/events/seminars/evolutionary-game-theory-with-payoff-fluctuations/

1 0

Master Thesis on Wednesday, Oct 27, 2021 at 09:30
by semfis＠ifisc.uib-csic.es 22 Oct '21

22 Oct '21

Master Thesis on Wednesday Oct 27, 2021 at 09:30 Place: Zoom Seminar Title: Evolutionary Game Theory with payoff fluctuations: Bet-Hedging Speaker: Rubén Calvo Ibáñez, IFISC/UIB Abstract: This thesis focuses on bet-hedging in changing evolutionary games. We study games with environmental fluctuations and quantify the effect that the distribu- tion of fitness of a given strategy has on its success. In general, it is an accepted idea in the field of biology, that reducing the variance of this distribution, even at the expense of reducing its mean, can be beneficial. We model this scenario as a simple game having stochastic fluctuations in the payoff matrix, and an- alyze it in detail. Our simulation results confirm the existence of bet-hedging strategies, in general, for a wide range of distributions. However, the particu- lar asymmetry that the distribution has is seen to play an important role. We were able to demonstrate that there exist new types of bet-hedging when the distribution is negatively skewed. These are characterized by an increase in the variance of fitness fluctuations leading to higher success, even at the expense of having to reduce mean fitness. Jury: Pere Colet, Manuel Matias,&nbsp;Tobias Galla Advisor: Tobias Galla Zoom:&nbsp;https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09 Url event:https://ifisc.uib-csic.es/en/events/seminars/evolutionary-game-theory-with-payoff-fluctuations/

1 0

Reminder: Master Thesis, TODAY - Friday, Oct 22, 2021 at 12:00
by semfis＠ifisc.uib-csic.es 22 Oct '21

22 Oct '21

Master Thesis on Friday Oct 22, 2021 at 12:00 Place: Zoom Seminar Title: Spontaneous synchronization in spin systems Speaker: Lucía Rodrigo Bort, IFISC Abstract: This Master Thesis is focused on the study of the fundamental properties that a system must possess to be able to show synchronization in the quantum regime. In the article &ldquo;Synchronizing the Smallest Possible System&rdquo; [1] it was claimed that quantum two-level systems are not a good candidate, as they lack a limit cycle and cannot be seen as self-sustained oscillators, therefore the smallest system to be a three-level system. Our intention is to raise objections to this idea using two main arguments. First, we start ques- tioning the definition of the limit cycle proposed in this article, as well as the premise that only systems with a limit cycle can be synchronized. And secondly, we point out that the quantum features observed in the article &ldquo;Synchronization and Entanglement Generation&rdquo; [2], in which the same authors submit the spontaneous synchronization between two coupled spins s = 1 systems, can be replicated with a pair of spin s = 1/2 systems, emphasizing the similarity between both systems. We&nbsp;analyse the evolution of the spin observables using different measures of synchronization. [1]&nbsp;Roulet and&nbsp;Bruder,&nbsp;Phys.&nbsp;Rev. Lett., 121, 053601 (2018) [2]&nbsp;Roulet and&nbsp;Bruder, Phys. Rev. Lett., 121, 063601 (2018) Jury: Rosa L&oacute;pez, Miguel C. Soriano, Gian Luca Giorgi Advisors: Gian Luca Giorgi, Roberta Zambrini Zoom:&nbsp;&nbsp;<a href="https://uibuniversitat.zoom.us/j/88073638068">https://uibuniversitat.zoom.us/j/88073638068</a> Url event:https://ifisc.uib-csic.es/en/events/seminars/spontaneous-synchronization-in-spin-systems/

2 1

Reminder: Talk, TODAY - Thursday, Oct 21, 2021 at 11:30
by semfis＠ifisc.uib-csic.es 21 Oct '21

21 Oct '21

Talk on Thursday Oct 21, 2021 at 11:30 Place: IFISC Seminar Room Title: Data-driven classification of animal trajectories Speaker: Jorge Medina, IFISC (CSIC-UIB) Abstract: Data science is revolutionizing our understanding in a wide variety of fields, such as speech recognition, autonomous driving and protein folding. This has been possible due to dramatic improvements in computational power and data collection and storage. In the case of animal movement, thanks to the development and deployment of tracking devices during the past decades, there exists now a considerable number of animal trajectories, susceptible of being analyzed via data-driven methods. However, its potential remains generally unexplored under these novel techniques. Our goal will be to assess the performance and adequateness of data-based tools in marine animal movement. Here, we will consider the species classification task as a proxy. Specifically, we will evaluate the effectiveness of several algorithms representative of the state of the art, as well as propose modifications in the input that could improve their ability to gain insight from the data. Url event:https://ifisc.uib-csic.es/en/events/seminars/data-driven-classification-of-animal-trajectories/

1 0

Reminder: IFISC Seminar, TODAY - Wednesday, Oct 20, 2021 at 14:30
by semfis＠ifisc.uib-csic.es 20 Oct '21

20 Oct '21

IFISC Seminar on Wednesday Oct 20, 2021 at 14:30 Place: Zoom Seminar Title: Does learning converge to equilibrium in generic games? Speaker: Marco Pangallo, Sant’Anna School of Advanced Studies, Pisa Abstract: Game theory models&nbsp;interacting biological and social systems. In a repeated game, players may converge or their dynamics oscillate. If the system is not designed to converge, which of these two behaviors can we expect?&nbsp;I will give an overview of research that addresses this question by studying convergence to equilibrium in generic games. I will consider simple 2-player, 2-action games, where players learn through Experience-Weighted Attraction learning. We are able to characterize convergence in&nbsp;the space of parameters and games. Games with a &ldquo;best-reply cycle&rdquo;, such as Matching Pennies, are the ones in which convergence is less likely. Further,&nbsp;I will show that the frequency of best-reply cycles predicts convergence of six learning rules in 2-player games with an arbitrary number of actions, when games are sampled at random given constraints.The larger the number of actions, or the more anti-correlated the payoffs, the more best-reply cycles become dominant, and convergence becomes less likely. Finally, I consider games with an arbitrary number of players and with a network structure. I&nbsp;show that more players and more dense networks increase the importance of best-reply cycles, making convergence unlikely. Overall, these results indicate that in many cases equilibrium is an unrealistic assumption, and one must explicitly model the dynamics of learning. <a href="https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09">https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09</a> Url event:https://ifisc.uib-csic.es/en/events/seminars/does-learning-converge-to-equilibrium-in-generic-g/

1 0

Reminder: Master Thesis, TODAY - Wednesday, Oct 20, 2021 at 16:30
by semfis＠ifisc.uib-csic.es 20 Oct '21

20 Oct '21

Master Thesis on Wednesday Oct 20, 2021 at 16:30 Place: Zoom Seminar Title: A normal form for excitable vegetation dynamics Speaker: Carles Martorell, IFISC (CSIC-UIB) Abstract: Master thesis presentation. In this work we study excitable dynamics related with vegetation ecosystems. In particular, we are interested in a model of Posidonia oceanica seagrass which motivates the dynamical system studied in this project. This model introduces an autotoxicity mechanism, by means of sulfide production due to decomposition of the seagrass, which plays a negative feedback role in the dynamics. In this way, this model can display interesting and sophisticated phenomena such as spatio-temporal patterns. From a Dynamical Systems point of view, all these phenomena are related with a rich bifurcation diagram. We propose a simple dynamical system which presumably reproduces such bifurcation diagram. In particular, we pay attention on the excitable region appearing close to the Takens-Bugdanov point. We study the existence of excitable travelling pulses in this region when a diffusion term is added in the spatially extended model. Jury: Pere Colet, Manuel A. Mat&iacute;as, and D. Gomila Join Zoom Meeting https://uibuniversitat.zoom.us/j/84856041678 &nbsp; Url event:https://ifisc.uib-csic.es/en/events/seminars/a-normal-form-for-excitable-vegetation-dynamics/

1 0

IFISC Seminar on Wednesday, Oct 20, 2021 at 14:30
by semfis＠ifisc.uib-csic.es 15 Oct '21

15 Oct '21

IFISC Seminar on Wednesday Oct 20, 2021 at 14:30 Place: Zoom Seminar Title: Does learning converge to equilibrium in generic games? Speaker: Marco Pangallo, Sant’Anna School of Advanced Studies, Pisa Abstract: <a href="https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09">https://us02web.zoom.us/j/83829318876?pwd=Z2pqbUtIMEV3NUQvU0hpakp0NGtsUT09</a> Url event:https://ifisc.uib-csic.es/en/events/seminars/does-learning-converge-to-equilibrium-in-generic-g/

1 0

Master Thesis on Wednesday, Oct 20, 2021 at 16:30
by semfis＠ifisc.uib-csic.es 15 Oct '21

15 Oct '21

Master Thesis on Wednesday Oct 20, 2021 at 16:30 Place: Zoom Seminar Title: A normal form for excitable vegetation dynamics Speaker: Carles Martorell, IFISC (CSIC-UIB) Abstract: Master thesis presentation. In this work we study excitable dynamics related with vegetation ecosystems. In particular, we are interested in a model of Posidonia oceanica seagrass which motivates the dynamical system studied in this project. This model introduces an autotoxicity mechanism, by means of sulfide production due to decomposition of the seagrass, which plays a negative feedback role in the dynamics. In this way, this model can display interesting and sophisticated phenomena such as spatio-temporal patterns. From a Dynamical Systems point of view, all these phenomena are related with a rich bifurcation diagram. We propose a simple dynamical system which presumably reproduces such bifurcation diagram. In particular, we pay attention on the excitable region appearing close to the Takens-Bugdanov point. We study the existence of excitable travelling pulses in this region when a diffusion term is added in the spatially extended model. Jury: Pere Colet, Manuel A. Mat&iacute;as, and D. Gomila Join Zoom Meeting https://uibuniversitat.zoom.us/j/84856041678 &nbsp; Url event:https://ifisc.uib-csic.es/en/events/seminars/a-normal-form-for-excitable-vegetation-dynamics/

1 0

Talk on Thursday, Oct 21, 2021 at 11:30
by semfis＠ifisc.uib-csic.es 15 Oct '21

15 Oct '21

Talk on Thursday Oct 21, 2021 at 11:30 Place: IFISC Seminar Room Title: Data-driven classification of animal trajectories Speaker: Jorge Medina, IFISC (CSIC-UIB) Abstract: Data science is revolutionizing our understanding in a wide variety of fields, such as speech recognition, autonomous driving and protein folding. This has been possible due to dramatic improvements in computational power and data collection and storage. In the case of animal movement, thanks to the development and deployment of tracking devices during the past decades, there exists now a considerable number of animal trajectories, susceptible of being analyzed via data-driven methods. However, its potential remains generally unexplored under these novel techniques. Our goal will be to assess the performance and adequateness of data-based tools in marine animal movement. Here, we will consider the species classification task as a proxy. Specifically, we will evaluate the effectiveness of several algorithms representative of the state of the art, as well as propose modifications in the input that could improve their ability to gain insight from the data. Url event:https://ifisc.uib-csic.es/en/events/seminars/data-driven-classification-of-animal-trajectories/

1 0