Graduate

Have you ever wondered how many samples Q-learning needs to learn a good policy? Is epsilon-greedy a good exploration strategy for reinforcement learning, or is there a better alternative? Although these questions are fundamental, they are not well understood even in the basic scenario with finitely many states and actions. In this talk, I will present our recent effort to answer both of the above questions with a provably sample-efficient version of Q-learning. This paper won the best paper award in ICML 2018 workshop "Exploration in RL."

TBA

The seminars are scheduled for 1:30 pm on the dates listed above. The speakers are renowned scholars or industry experts in power and energy systems. We believe they will bring novel insights and fruitful discussions to Stanford. This seminar is offered as a 1 unit seminar course, CEE 272T/EE292T. Interested students can take this seminar course for credit by completing a project based on the topics presented in this course.

Yours sincerely,
Smart Grid Seminar Organization Team,

Ram Rajagopal, Associate Professor, Civil & Environmental Engineering, and Electrical Engineering
Sila Kiliccote, Managing Director of Grid Innovations, Bits & Watts 
Chin-Woo Tan, Director, Stanford Smart Grid Lab 
Yuting Ji, Postdoctoral Scholar, Civil and Environmental Engineering

TBA

The seminars are scheduled for 1:30 pm on the dates listed above. The speakers are renowned scholars or industry experts in power and energy systems. We believe they will bring novel insights and fruitful discussions to Stanford. This seminar is offered as a 1 unit seminar course, CEE 272T/EE292T. Interested students can take this seminar course for credit by completing a project based on the topics presented in this course.

Yours sincerely,
Smart Grid Seminar Organization Team,

Ram Rajagopal, Associate Professor, Civil & Environmental Engineering, and Electrical Engineering
Sila Kiliccote, Managing Director of Grid Innovations, Bits & Watts 
Chin-Woo Tan, Director, Stanford Smart Grid Lab 
Yuting Ji, Postdoctoral Scholar, Civil and Environmental Engineering

TBA

The seminars are scheduled for 1:30 pm on the dates listed above. The speakers are renowned scholars or industry experts in power and energy systems. We believe they will bring novel insights and fruitful discussions to Stanford. This seminar is offered as a 1 unit seminar course, CEE 272T/EE292T. Interested students can take this seminar course for credit by completing a project based on the topics presented in this course.

Yours sincerely,
Smart Grid Seminar Organization Team,

Ram Rajagopal, Associate Professor, Civil & Environmental Engineering, and Electrical Engineering
Sila Kiliccote, Managing Director of Grid Innovations, Bits & Watts
Chin-Woo Tan, Director, Stanford Smart Grid Lab
Yuting Ji, Postdoctoral Scholar, Civil and Environmental Engineering

California has the will, ambition, technology and legal requirement to decarbonize our energy sector by 2045. In the dozen years since passage of AB32, we have made great progress but we may be making some grave mistakes. In this discussion, Dr. Fine describes how distributed energy resources are presenting new opportunities in distribution resources, transmission and procurement planning, and market reforms that will determine if our clean energy future is one that is affordable for all.

Since the advent of optical microscopy, a flat glass microscope slide has been the standard surface upon which tissues, cells, and biomolecules are attached for observation. Recently, we have utilized the optically resonant properties of nanostructured photonic crystal (PC) surfaces to enable several new microscopy modalities where the nanostructure provides new forms of contrast for a wide variety of compelling applications. By designing PC surfaces with resonances that match the excitation and emission spectra of photon emitters such as fluorophores and quantum dots, PC enhanced fluorescence (PCEF) microscopy reduces the detection limits of any surface-based fluorescence assay. By generating spatial images of the PC resonant reflection intensity, we can selectively and dynamically visualize the cell-extracellular matrix interface during processes that include stem cell differentiation, cancer cell response to drugs, and chemotaxis – including the ability to observe the formation and evolution of cell membrane focal adhesion sites. We call this approach Photonic Resonator Outcoupler Microscopy (PROM), as we observe highly localized outcoupling of light from the PC that occurs due to scattering by dense regions in the cell membrane. By utilizing metallic nanoparticle tags with plasmon resonances that match the PC resonance, we obtain highly efficient coupling of light into nanometer-scale electromagnetic hotspots, that is accompanied by highly localized "quenching" of the PC reflection efficiency, which we are using for digital-resolution detection of miRNA biomarkers for cancer with 100 aM limits of detection, using an approach called Photonic Resonator Absorption Microscopy (PRAM). The seminar will describe the physical principles, nanostructure design/fabrication, instrumentation, and applications for nanostructure-enabled microscopy for disease diagnostics, personalized medicine, and life science research.

Machine learning and data analytics continue to expand the fourth industrial revolution and affect many aspects of our lives. This talk will explore machine learning applications in data-driven neuroscience and low-energy implementations of machine learning systems. Data-driven neuroscience can exploit machine learning approaches including deep learning to generate hypotheses associated with biomarkers for specific neuro-psychiatric disorders. In the first part, I will talk about use of machine learning to find biomarkers for epilepsy and adolescent mental disorders such as borderline personality disorder (BPD), using electroencephalogram (EEG) and functional magnetic resonance imaging (fMRI), respectively. In the second part of the talk, I will talk about approaches for energy-efficient implementations for both traditional machine learning and deep learning systems. I will talk about the roles of feature ranking and incremental-precision approaches to reduce energy consumption of traditional machine learning systems. I will then talk about our recent work on Perm-DNN based on permuted-diagonal interconnections in deep convolutional neural networks and how structured sparsity can reduce energy consumption associated with memory access in these systems.

REGISTRATION DEADLINE IS WEDNESDAY, OCTOBER 17, 2018. REGISTRATION IS REQUIRED. GRADUATE STUDENTS/POSTDOCS, PLEASE REGISTER.

Learn from other changemakers in STEM, and build your professional and academic network at WINS. The WISE Inspirations Network at Stanford (WINS) provides you the opportunity to engage with a diverse array of stand-out women in science and engineering, and connect with potential mentors, peers, and other colleagues.

The Fall 2018 WINS will feature Ellen Ochoa, PhD.

Ellen Ochoa served as the Director of NASA's Johnson Space Center, leading the human space flight enterprise for the nation. The first Latina in space on the shuttle Discovery, she has flown in space four times, leading onboard scientific activities, operating the robotic arm, and serving as flight engineer during the launch, rendezvous, and entry phases of the mission. She earned Ph.D. and M.S. degrees in Electrical Engineering at Stanford, following a B.S. in Physics from San Diego State University. A former Stanford trustee, Dr. Ochoa currently provides executive guidance and board service for a variety of organizations.

For the event, Ellen Ochoa will be in dicussion with Elizabeth Hadly, Paul S. and Billie Achilles Professor in Environmental Biology, Faculty Director of Jasper Ridge Biological Preserve, Senior Fellow at the Woods Institute for the Environment, Senior Fellow at the Stanford Center for Innovation in Global Health, and Professor, by courtesy, of Geological Sciences.

Elizabeth Hadly first became interested in understanding dynamics of the natural world after exploring the Smithsonian Institution and wild lands in the west as a child. Her research on the past, present and future environmental impacts on the ecology and evolution of animals has taken her from the arctic to the tropics and from sea level to the alpine of the Americas, Asia, and Africa. Professor Hadly and her lab use a myriad of laboratory and field tools including molecular genetics, fossils, isotopes, modeling, and animal population dynamics, and are committed to communicating to the public about global change. She is the co-author of Tipping Point for Planet Earth: How Close Are We to the Edge? with Tony Barnosky in 2016.

There is also a Leadership Dinner Opportunity. Please go to the Office of the Provost for Graduate Education event webpage for details.

Transport in strongly quantum systems is challenging to understand. I will describe a recently obtained bound on transport in terms of a characteristic quantum velocity (the Lieb-Robinson velocity) and the local thermalization time. This bound sheds some light on experiments in both condensed matter systems and ultracold atomic gases. At finite temperatures, a more powerful velocity is the so-called butterfly velocity, that is intimately related to quantum chaos. This velocity is still poorly understood; I will present some forthcoming results that constrain the temperature dependence of the butterfly velocity in terms of the underlying quantum scrambling of the system.

Optical imaging probes like otoscopes and laryngoscopes are essential tools used by doctors to see deep into the human body. Until now, they have been crucially limited to two-dimensional (2D) views of tissue lesions in vivo that frequently jeopardize their diagnostic usefulness. Depth imaging is critically needed in medical diagnostics because most tissue lesions manifest themselves as abnormal 3D structural changes. In this talk, I will talk our recent effort to develop three-dimensional (3D) plenoptic imaging tool that revolutionizes diagnosis with unprecedented sensitivity and specificity in the images produced. Particularly, I will discuss two plenoptic medical cameras, a plenoptic otoscope and a plenoptic laryngoscope, and their applications for in-vivo imaging.