Speaker: Melody Guan Title: Efficient Neural Architecture Search via Parameter Sharing Abstract We propose Efficient Neural Architecture Search (ENAS), a fast and inexpensive approach for automatic model design. In ENAS, a controller learns to discover neural network architectures by searching for an optimal subgraph within a large computational graph. The controller is trained with policy gradient to select a subgraph that maximizes the expected reward on the validation set. Meanwhile the model corresponding to the selected subgraph is trained to…
Author Archive
Future robots will probably be more intelligent than C3PO. Come find out why. Join us on Wednesday, January 17, 2018 in Curia II (WH2SW) at 12pm to discuss papers, challenges, and ideas in deep learning and artificial intelligence.
Deep Learning Seminar Friday, November 3 1:00 – 2:00pm Curia II (WH2SW) Hybrid classical/quantum algorithms for Fermion simulation and approximate combinatorial optimization Nicholas C. Rubin Rigetti Computing Hybrid classical/quantum algorithms have the characteristic of utilizing both classical and quantum computational resources to facilitate a computational advantage. This class of algorithms are likely candidates for near-term quantum computers because of their natural robustness to incoherent noise sources. In this talk we will introduce the general paradigm of hybrid classical quantum algorithms…
Deep Learning Journal Club! What is Deep Learning? What happens when Data comes alive? Join us to discuss papers, problems, and code for deep learning and artificial intelligence. Wednesday, October 18 12pm Racetrack (WH7X)
Interested in Deep Learning? Join us for an odyssey through papers, problems and code for deep learning and artificial intelligence. The Deep HEP journal club meets on the second Tuesday of every month, and is open to everyone interested. Please join us for the next meeting. March 14, 2017 12 p.m. Hornet’s Nest (WH8X) If you can’t make it in person, Join via Zoom! 473918307
Are you interested in deep learning? Join other physicists in a lab-wide group to share tools, skills, and resources. Let’s push the limits. We’ll share resources, projects we’re working on and discuss the potential for deep learning at Fermilab. For more information, please contact Alexander Radovic (a.radovic@gmail.com) or join us on slack at hepmachinelearning.slack.com When: Oct 26, 2016, 6pm Where: Fermilab Users Center