PhD Talk/Colloquia

Spring Semester

The spring colloquium is a one-hour format, designed for practicing longer conference presentations or job talks. Participants in the spring colloquium will be candidates for the Best PhD Colloquium award.

Please sign up here.

**Seminars and Talks will be held in Towne 337 on Wednesdays at 12:00 PM unless otherwise specified.**

Luiz Chamon
Wednesday, February 21st
"Strong Duality of Sparse Functional Optimization"

Learn more about this colloquium

Abstract: Signal processing is rich in inherently continuous applications, such as radar, MRI, and source localization, in which sparsity priors play a key role in obtaining state-of-the-art results. To cope with the infinite dimensionality and non-convexity of these estimation problems, they are typically discretized and solved by means of convex relaxations, e.g., using atomic norms. Although successful, this approach is not without issues. Discretization often leads to high dimensional, potentially ill-conditioned optimization problems. Moreover, due to grid mismatch and other coherence issues, a sparse signal in the continuous domain may no longer be sparse when discretized. Finally, performance guarantees for atomic norm relaxations hold under assumptions that may be hard to meet in practice. We address these issues by directly tackling the continuous problem cast as a sparse functional optimization program. We prove that these problems have no duality gap and show that they can be solved efficiently using duality and a stochastic gradient ascent-type algorithm. We illustrate the performance of this new approach on a line spectral estimation problem.

Bio: Luiz Chamon received the B.Sc. and M.Sc. degree in electrical engineering from the University of São Paulo, Brazil, in 2011 and 2015. During this period, his research involved adaptive filtering, acoustic MIMO equalization, and low complexity decimation/interpolation structures. In 2009, he participated in an undergraduate exchange with the Masters in Acoustics of the École Centrale de Lyon, France. He also worked on acoustical design and electronics for signal processing projects and provided statistical consulting on topics such as psychology and ergonomics. He is currently a Ph.D. candidate of the Department of Electrical and Systems Engineering at the University of Pennsylvania under the supervision of Prof. Alejandro Ribeiro. His research interests include signal processing, discrete optimization, statistics, and control.

Mark Eisen
Wednesday, February 28th
"Learning in Wireless Autonomous Systems"

Learn more about this colloquium

Abstract: This work considers the task of learning optimal power allocation policies in a wireless control systems over an unknown time-varying non-stationary channel. The goal is to maximize control performance of a set of independent control systems by allocating transmitting power within a fixed budget. Since the channel's time-varying distribution is unknown, samples of the channel are taken at every epoch. By reverting the resulting stochastic optimization problem in its Lagrange dual domain, we demonstrate that it takes the equivalent form of minimizing a certain empirical risk measure, a well-studied problem in machine learning. Newton's method is used to quickly learn approximately optimal power allocation policies over the sampled dual function as the channel evolves over time over windows of epochs. The quadratic convergence rate of Newton is used to establish, under certain conditions on the sampling size and rate of channel variation, an instantaneous learning and tracking of these optimal policies. Numerical simulations demonstrate the effectiveness of the learning algorithm on a low-dimensional wireless control problem.

Bio: Mark Eisen received the B.Sc. degree in electrical engineering from the University of Pennsylvania, Philadelphia, USA in 2014. He is now working towards his PhD in the Department of Electrical and Systems Engineering at the University of Pennsylvania. His research interests include distributed optimization and machine learning. In the summer of 2013, he was a research intern at the Institute for Mathematics and its Applications at the University of Minnesota, Minneapolis, MN. Mr. Eisen was awarded Outstanding Student Presentation at the 2014 Joint Mathematics Meeting, as well as the recipient of the 2016 Penn Outstanding Undergraduate Research Mentor Award.

Bhoram Lee
Wednesday, March 28th
"Probabilistic Online Learning of Appearance and Structure for Robotics"

Learn more about this colloquium


Abstract: A robotic system can be characterized by its physical interaction with environments via continuous sensing, acting, and learning. Visual and spatial sensing of such a system plays a critical role in the perception of the world and accounts for a rich source of learning. Although recent advances in machine perception have presented unprecedented performance in some areas, there still exist challenges in various aspects.

First of all, most current learning approaches in machine vision depend on fragments of exemplars prepared by humans. Ideally, a robotic system is required to gather information and keep growing knowledge on the fly without constant external aids. One way to implement the self-learning is to take advantage of the structure naturally available in environments which results in correlated sensations of different sensory modalities. The first part of this talk introduces a probabilistic online self-learning framework to leverage structural priors in order to alleviate the dependency in robotic visual learning.

Another challenge in robotics is its spatial understanding, which still largely requires more research. While point or grid based representations are currently being employed for practical conveniences, these methods suffer from discretization and disconnected spatial information. On the other hand, Gaussian Processes (GP) has recently gained attention as an alternative to represent spatial structures continuously and probabilistically. The seamless expression of structures is not only close to how humans perceive the world, but also invaluable for other robotic tasks such planning and control. The second part of this talk suggests an online framework for building continuous spatial structure using GP.

Bio: Bhoram Lee is currently a PhD candidate at GRASP Lab, University of Pennsylvania. Before coming to Penn, she worked at SAIT (Samsung Advanced Institute of Technology) from 2007 to 2013 as a researcher. Bhoram received B.S. in mechanical and aerospace engineering in 2005 and M.S. in aerospace engineering in 2007 from Seoul National University, Korea. Her previous research experience includes visual navigation of UAVs, sensor fusion, and mobile user interfaces. Bhoram’s current interest includes 3D vision, machine learning, and general robotics and is focused on online learning for robot vision.


Fernando Gama
Wednesday, April 4th
"Convolutional Neural Networks on Graph Signals"

Learn more about this colloquium

Abstract: Convolutional neural networks (CNNs) are being applied to an increasing number of problems and fields due to their superior performance in classification and regression tasks. Since two of the key operations that CNNs implement are convolution and pooling, this type of networks is implicitly designed to act on data described by regular structures such as images or time series. Motivated by the recent interest in processing signals defined in irregular domains, we elaborate a comprehensive framework for enabling convolutional neural networks to fully operate on graph signals. The convolution operation is straightforwardly replaced by graph filtering. For extending the pooling stage, we decouple it in two operations: a local nonlinearity creating neighborhood summaries, and a downsampling operation. Then, motivated by existing methods for sampling graph signals, we propose two pooling methods, namely selection pooling and aggregation pooling. In the first case, we select a subset of nodes that pool te features from their neighbors, and then, for the next layer, we zero-pad the not-selected nodes and filter again to obtain more features; then, select a smaller subset of nodes and so on. In the aggregation pooling CNN, we select a single node and sequentially obtain information from neighborhoods to create a regular structure on the data that incorporates the underlying irregular support. Once the data has a regular structure, we proceed to apply a traditional CNN, thereby relating neighborhood information. Architectures in the proposed framework are tested in a synthetic example of source localization, as well as a classification problem in the 20NEWS dataset.

Bio: Fernando Gama is a 4th-year Ph.D. student at the Electrical and Systems Engineering department of the University of Pennsylvania. He graduated as Electronic Engineer from the University of Buenos Aires in 2013, and got a Fulbright Scholarship for international students to start his Ph.D. in 2014. He got his M. A. in Statistics from the University of Pennsylvania in 2017. His research interests are in the fields of information processing, machine learning and neural networks, applied to network data.

Nick Watkins
Wednesday, April 18th
"Predictive Control Methods for Epidemic Processes"

Learn more about this colloquium

Abstract: There are many examples of complex, interconnected systems in our modern world: computer networks, robotic swarms, and human societies are frequently studied examples. Epidemic processes are a popular formal mathematical model for such systems, wherein the spread of a certain status (e.g. infection with a particular disease) evolves according to given stochastic (i.e. chance driven) rules. Despite the inherent importance of understanding how to control the evolution of such processes, most research on the control of epidemic processes to date provide performance guarantees only with respect to mean-field approximations. This is an important restriction, because even if epidemic processes were themselves perfect models of real-world complex systems, it is unclear if the mean-field approximation used would allow the controllers developed to be effective in practice. In this talk, we show how concepts from control theory can be used to build predictive controllers which provide rigorous performance guarantees on the epidemic processes themselves, thus making it clear that if epidemic processes are good models for complex, interconnected systems, such systems can be well controlled.

Bio: Nick Watkins is currently a Ph.D. student in the Electrical and Systems Engineering program at the University of Pennsylvania. He graduated summa cum laude with a B.S. in Electrical Engineering from Wilkes University in 2013, as the top student in his class. His current research interests are focused on the control and optimization of stochastic systems, with projects focusing on controlling spread in epidemic processes and managing resources efficiently in energy harvesting devices.

Santiago Paternain
Wednesday, April 25th
"Stochastic Policy Gradient Ascent in Reproducing Kernel Hilbert Spaces"

Learn more about this colloquium

Abstract: In this talk, we consider the problem of policy optimization in the context of reinforcement learning. In order to avoid discretization, we select the optimal policy to be a continuous function belonging to a Reproducing Kernel Hilbert Space (RKHS) which maximizes an expected discounted reward (EDR). We design a policy gradient algorithm (PGA) in this context, deriving the gradients of the functional EDR and learning the unknown state transition probabilities on the way. In particular, we propose an unbiased stochastic approximation for the gradient that can be obtained in a finite number of steps. This unbiased estimator is the key enabler for a novel stochastic PGA, which provably converges to a critical point of the EDR. However, the RKHS approach increases the model order per iteration by adding extra kernels, which may render the numerical complexity prohibitive. To overcome this limitation, we prune the kernel dictionary using an orthogonal matching pursuit procedure and prove that the modified method keeps the model order bounded for all iterations while ensuring convergence to a neighborhood of the critical point.

Bio: Santiago Paternain received the B.Sc. degree in electrical engineering from Universidad de la Republica Oriental del Uruguay, Montevideo, Uruguay in 2012. Since August 2013, he has been working toward the Ph.D. degree in the Department of Electrical and Systems Engineering, University of Pennsylvania. His research interests include optimization and control of dynamical systems. Mr. Paternain received the 2017 CDC best paper award.

 

Fall Semester

The Ph.D. colloquium has added a new, shorter format: the Ph.D. Talk! Ph.D. Talks are scheduled throughout the fall. The talks are 20 minutes and informal, with the purpose of letting students discuss their research, get feedback and ideas, and share resources.

Santiago Paternain
Wednesday, October 11th
"A Second Order Method for Nonconvex Optimization"

Learn more about this talk

Talk Info: Machine learning problems such as neural network training, tensor decomposition, and matrix factorization, require local minimization of a nonconvex function. This local minimization is challenged by the presence of saddle points, of which there can be many and from which descent methods may take an inordinately large number of iterations to escape. In this talk, we present a second-order method that modifies the update of Newton's method by replacing the negative eigenvalues of the Hessian by their absolute values and uses a truncated version of the resulting matrix to account for the objective function's curvature. The method is shown to escape saddles exponentially with base 1.5 regardless of the condition number of the problem. Adding classical properties of Newton's method, the paper proves convergence to a local minimum with high probability after a number of iterations that is logarithmic in the target accuracy.

Bio: Santiago Paternain received the B.Sc. degree in electrical engineering from Universidad de la Republica Oriental del Uruguay, Montevideo, Uruguay in 2012. Since August 2013, he has been working toward the Ph.D. degree in the Department of Electrical and Systems Engineering, University of Pennsylvania. His research interests include optimization and control of dynamical systems.

Mark Eisen
Wednesday, October 18th
"A Decentralized Primal-Dual Quasi-Newton Method for Consensus Optimization"

Learn more about this talk

Talk Info: This work considers consensus optimization problems where each node of a network has access to a different summand of an aggregate cost function. Nodes try to maximize the aggregate cost function, while they exchange information only with their neighbors. We modify the augmented Lagrangian method to incorporate a curvature correction inspired by the Broyden-Fletcher-Goldfarb-Shanno (BFGS) quasi-Newton method. The resulting primal-dual BFGS method is a fully decentralized algorithm in which nodes approximate curvature information of themselves and neighbors for both the primal and dual variables of the augmented Lagrangian through the satisfaction of a secant condition. The developed algorithm is of interest in consensus problems that are not well conditioned, making first order decentralized methods ineffective, and in which second order information is not readily available, making decentralized second order methods infeasible. We establish a linear convergence rate to the exact solution of the consensus problem and performance advantages relative to alternative decentralized algorithms are shown numerically.

Bio: Mark Eisen received the B.Sc. degree in electrical engineering from the University of Pennsylvania, Philadelphia, PA, USA, in 2014. He is currently working toward the Ph.D. in the Department of Electrical and Systems Engineering, University of Pennsylvania. His research interests include distributed optimization and machine learning. In the summer of 2013, he was a research intern in the Institute for Mathematics and its Applications, University of Minnesota, Minneapolis, MN, USA. He received the Outstanding Student Presentation at the 2014 Joint Mathematics Meeting, as well as the 2016 Penn Outstanding Undergraduate Research Mentor Award.

Andreea Alexandru
Wednesday, October 18th
"Privacy Preserving Cloud-Based Quadratic Optimization"

Learn more about this talk

Talk Info: In the Internet of Things setup, cloud-outsourced computations are ubiquitous, because of the low computation, battery and storage requirements of the participating devices. Due to the increasing number of cyberattacks, privacy infringements and financial interests arising from owning private data, it is unrealistic to assume that the cloud does not try to take advantage of the users' data. The most common framework in which multi-party computation is performed is the semi-honest model, which, intuitively, describes rival parties that collaborate to achieve a common goal. Under this setup, we wish to develop protocols that satisfy cryptographic security, i.e., no party can infer anything about the private data of other parties. More specifically, we address optimization problems, which lie at the core of control applications, such as state estimation, model predictive control etc. In this talk, we propose a protocol for privately solving constrained quadratic optimization problems with sensitive data. The problem encompasses the private data of multiple agents and is outsourced to an untrusted server. We present an interactive protocol that achieves the solution by making use of partially homomorphic cryptosystems to securely effectuate computations.

Bio: Andreea Alexandru received the B.Sc. degree in Automatic Control and Systems Engineering from “Politehnica” University of Bucharest, Romania, in 2015. She is currently in her third year of Ph.D. program in the Department of Electrical and Systems Engineering, University of Pennsylvania, working with prof. George Pappas and prof. Ali Jadbabaie. Her research interests lie in the security of control systems, involving both cryptographic and information-theory tools.

David Hopper
Wednesday, October 25th
"Amplified Sensitivity of Nitrogen-Vacancy Center Sensors with All-Optical Charge Readout"

Learn more about this talk

Talk Info: The nitrogen-vacancy (NV) center in diamond is a solid-state qubit and nanoscale sensor for applications including single-molecule nuclear magnetic resonance and dynamic electrochemical potential sensing with nanoscale resolution. These demonstrations are possible due to the NV’s photoluminescence dependence on changes in the spin or charge state of the qubit, which provides an optical indicator of the external environment. Despite these impressive demonstrations, ideal sensing platforms such as nanodiamonds exhibit poor signal-to-noise ratios for spin and charge measurements, which prevent the wide adoption of these sensing capabilities. In this talk, I will discuss recent work from our group on circumventing these technological hurdles by leveraging full control over the NV’s spin, orbital, and charge dynamics in nanodiamonds. Critical to our method is the development of a high signal-to-noise ratio, all-optical charge readout protocol along with a means for efficiently correlating a spin state with a charge distribution. I will conclude by discussing how these methods can improve current state-of-the-art NV sensors.

Bio: David Hopper earned his bachelor’s degree in Physics with Honors from The Pennsylvania State University in 2014. He is currently pursuing his PhD in Physics at the University of Pennsylvania with Professor Lee Bassett as part of the physics graduate group. He is broadly interested in semiconductor quantum dynamics and quantum information science, with a current focus on improving the readout capabilities of the nitrogen-vacancy center in diamond.

Mahyar Fazlyab
Wednesday, October 25th
"Analysis of optimization algorithms via Integral Quadratic Constraints: Non-strongly convex problems"

Learn more about this talk

Talk Info: In this work, we develop a unified framework, based on robust control theory and semidefinite programming, to analyze the performance of iterative first-order optimization algorithms, as well as their continuous-time counterparts. Our starting point is to represent these algorithms as linear dynamical systems interconnected with a nonlinear component. We then propose a family of time-dependent nonquadratic Lyapunov functions that are particularly useful for establishing arbitrary (exponential or subexponential) convergence rates. Using Integral Quadratic Constraints (IQCs) from robust control theory to describe the class of nonlinearities in the interconnection, we derive sufficient conditions for the Lyapunov stability of these algorithms in terms of Linear Matrix Inequalities (LMIs), whose size is independent of the problem dimension. We show how the developed LMI-based framework unifies the convergence analysis by studying several algorithms, namely, the gradient method, the Nesterov's accelerated method, proximal algorithms, and their accelerated variants. We perform the analysis for both strongly-convex and convex settings, where we expect exponential and subexponential convergence rates, respectively.

Bio: Mahyar received the B.Sc. and M.Sc. degree in Mechanical engineering from Sharif University of Technology, Tehran, Iran. Since 2013, he has been working towards the Ph.D. degree in Electrical and Systems Engineering at University of Pennsylvania. His research interests include the analysis, optimization, and control of dynamical systems.

Luiz Chamon
Wednesday, November 1st
"Approximate Supermodularity Bounds for Experimental Design"

Learn more about this talk

Talk Info: This work provides performance guarantees for the greedy solution of experimental design problems. In particular, it focuses on A- and E-optimal designs, for which typical guarantees do not apply since the mean-square error and the maximum eigenvalue of the estimation error covariance matrix are not supermodular. To do so, it leverages the concept of approximate supermodularity to derive non-asymptotic worst-case suboptimality bounds for these greedy solutions. These bounds reveal that as the SNR of the experiments decreases, these cost functions behave increasingly as supermodular functions. As such, greedy A- and E-optimal designs approach (1-1/e)-optimality. These results reconcile the empirical success of greedy experimental design with the non-supermodularity of the A- and E-optimality criteria.

Bio: Luiz Chamon received the B.Sc. and M.Sc. degree in electrical engineering from the University of São Paulo, Brazil, in 2011 and 2015. During this period, his research involved adaptive filtering, acoustic MIMO equalization, and low complexity decimation/interpolation structures. In 2009, he participated in an undergraduate exchange with the Masters in Acoustics of the École Centrale de Lyon, France. He also worked on acoustical design and electronics for signal processing projects and provided statistical consulting on topics such as psychology and ergonomics. He is currently a Ph.D. candidate of the Department of Electrical and Systems Engineering at the University of Pennsylvania under the supervision of Prof. Alejandro Ribeiro. His research interests include signal processing, discrete optimization, statistics, and control.

Anastasios Tsiamis
Wednesday, November 8th
"State-Secrecy Codes for Networked Linear Systems"

Learn more about this talk

Talk Info: A fundamental vulnerability of the Internet of Things is eavesdropping attacks, especially when the underlying medium of communication is a wireless one. In this talk, we focus on eavesdropping attacks in the context of networked linear dynamical systems.
An authorized user estimates the state of a linear plant, based on the packets received from a sensor, while the packets may also be intercepted by the eavesdropper. Our goal is to design a coding scheme at the sensor, which encodes the state information, in order to impair the eavesdropper's estimation performance, while enabling the user to successfully decode the sent messages. We introduce a novel class of codes, termed State-Secrecy Codes, which are fast and suitable for real-time dynamical systems.
They exploit acknowledgment signals from the user, the system's process noise, the channel randomness and the system's dynamics. Under minimal conditions, State-Secrecy Codes make the eavesdropper's state estimation error to be large, while the user's state estimation error is optimal. These conditions only require that at some time, the eavesdropper fails to intercept the respective packet, while the user receives it.

Bio: Anastasios Tsiamis received the Diploma degree in Electrical and Computer Engineering from the National Technical University of Athens, Greece, in 2014. Currently, he is a Ph.D. student in the Department of Electrical and Systems Engineering, University of Pennsylvania, working with prof. George Pappas.
His research interests include control systems security and networked control systems.

Dushyant Sahoo
Wednesday, November 15th
"GPU Accelerated Extraction of Sparse Granger Causality Patterns"

Learn more about this talk

Talk Info: Resting-state functional MRI, which provides a means to estimate the entire brain functional connectivity, has recently received a considerable amount of interest. This imaging method captures time series and is increasingly being used to study dynamics in the human brain. In this work, we propose to analyze fMRI dynamics by extracting Granger Causality patterns which are shared across subjects. This approach allows capturing organization of individual brain while extracting population-wide causality patterns which are more robust with respect to noise. A non-convex non-smooth optimization problem is framed for extracting causality networks and the problem is solved using Proximal Alternating Linearized Minimization. We introduce a fast and scalable method implemented on GPU for the extraction of shared causality patterns.

Bio: Dushyant Sahoo received his B.Tech. degree in Electrical Engineering and M.Tech. degree in Information and Communication Technology from Indian Institute of Technology, Delhi, India in 2016. He is currently in his first year of Ph.D. in the Department of Electrical and Systems Engineering, University of Pennsylvania, working with Prof. Christos Davatzikos. His research interests lie in analyzing fMRI data, focussing on finding causal networks in the human brain.

Cassiano Becker
Wednesday, November 29th
"Network Design for Controllability Metrics"

Learn more about this talk

Talk Info: In this talk, we address the constrained design of linear dynamics to improve system control performance, which can be measured as a function of the controllability Gramian. In contrast with the problem of deployment of actuation capabilities to achieve a specified control performance, we seek to change the dynamics of linear systems while considering the deployed actuation mechanisms. Specifically, we consider spectral properties of the 'infinite' controllability Gramian as control performance metrics, and apply constrained (i.e., bounded) perturbations in the system's parameters while respecting its structure. We show that two different (yet related) re-design problems for control enhancement can be cast as bilinear or linear matrix equality problems. Lastly, we propose different strategies to obtain the solution of these problems, and assess their performance in the context of multi-agent networks in the leader-follower setup.

Bio: Cassiano O. Becker is a Ph.D. student in the Department of Electrical and Systems Engineering at the University of Pennsylvania. He received an M.Sc. in Telecommunications (with distinction) from the University College London, an M.S. in Electrical Engineering from the State University of Campinas and a B.Eng. in Electrical Engineering from the Federal University of Rio Grande do Sul, Brazil. Before starting his doctoral studies, he worked as systems engineer, software developer and project manager in technology companies such as Harris Corporation and Siemens. His research interests include applications of network models, dynamical systems and machine learning to neuroscience, brain-computer interfaces and other cyber-physical systems.

Nick Watkins
Wednesday, November 29th
"Battery Management for Control Systems with Energy Harvesting Sensors"

Learn more about this talk

Talk Info: Energy harvesting devices - those which are capable of restoring their charge through interactions with their environment - are likely to be a key driving force in the development of future internet-of-things technology. In this talk, we study the problem of computing the minimum battery capacity required to stabilize a scalar plant communicating with an energy harvesting sensor over a wireless communication channel. We prove that a particular greedy battery management policy suffices to stabilize the plant, and demonstrate that stability of the system under the greedy policy can be checked by a linear program. Moreover, we show that a critical battery capacity exists, below which no policy can stabilize the system, which itself can be computed by solving a sequence of linear programs which grows logarithmically with respect to the maximum allowed storage capacity. The first of these results address an open question pertaining to the stability of energy harvesting control systems. The last allows us to efficiently compute the smallest battery capacity required to stabilize a given system, which addresses a problem of importance when device size or cost are significant concerns.

Bio: Nick Watkins is currently a Ph.D. student in the Electrical and Systems Engineering program at the University of Pennsylvania. He graduated summa cum laude with a B.S. in Electrical Engineering from Wilkes University in 2013, as the top student in his class. His current research interests are focused on the control and optimization of stochastic systems, with projects focusing on mitigating spread in epidemic processes and managing resources efficiently in energy harvesting devices.

Farshid Ashtiani
Wednesday, December 13th
"Towards Integrated Optical Frequency Synthesizer"

Learn more about this talk

Talk Info: Electrical frequency synthesizer integrated circuits have been extensively used in almost every communication and sensing systems. For instance, in every smartphone, these electronic chips play an integral role in receiving signals/data at different radio frequencies corresponding to different wireless communication standards that the device supports such as WiFi and LTE. Similarly, in optical domain, optical frequency synthesizers have many applications such as optical communication, spectroscopy, and frequency metrology. However, demonstrations of optical synthesizer have been mostly limited to bench-top systems rather than integrated implementation. Therefore, high cost, large size, and high power consumption have significantly limited their large-scale deployment. In this work, we present an integrated electro-optical phase-locked loop as one of the main building blocks of an optical synthesizer as well as demonstration of optical frequency synthesis. Benefitting from a standard silicon-on-insulator CMOS fabrication technology, this work is an important step towards implementation of low-cost, low-power, and reliable integrated optical frequency synthesizers.

Bio: Farshid Ashtiani received his B.Sc. and M.Sc. in electrical engineering from Sharif University of Technology in 2011 and 2013, respectively. Since 2014, he is pursuing his Ph.D. in Electrical and Systems Engineering at the University of Pennsylvania under supervision of Prof. Firooz Aflatouni. His research interest includes electronic-photonic integrated circuits and systems design.

Mohamad H. Idjadi
Wednesday, December 13th
"Integration of a laser stabilization system on a standard CMOS process"

Learn more about this talk

Talk Info: Low noise stable lasers have far-reaching applications in spectroscopy, communication, metrology and basic science. The laser’s frequency is determined by the optical length of the cavity in which the electromagnetic energy is confined. Despite all the improvements made in laser fabrication technology, laser cavity properties are extremely sensitive to environmental fluctuations such as temperature variation and vibration. These fluctuations affect the optical length of the cavity and inevitably, the frequency of the laser fluctuates. Many different laser stabilization techniques have been recently explored; the Pound−Drever−Hall (PDH) laser stabilization technique is widely used to stabilize different types of lasers. We have demonstrated the first integrated PDH system that can stabilize a low-cost laser to realize a compact inexpensive light source, which can ultimately impact many fields of science and engineering.

Bio: Mohamad H. Idjadi received his B.Sc. degree in electrical engineering from University of Tehran, Tehran, Iran in 2014. He has been working towards the Ph.D. degree in Electrical and Systems Engineering at the University of Pennsylvania since 2014. His research focuses on silicon photonic integrated systems and high-speed integrated circuits with applications in imaging, communication, Radar, and LIDAR.