Timezone: »

 
Poster
Superposition of many models into one
Brian Cheung · Alexander Terekhov · Yubei Chen · Pulkit Agrawal · Bruno Olshausen

Thu Dec 12 05:00 PM -- 07:00 PM (PST) @ East Exhibition Hall B + C #26
in Algorithms -- Online Learning »

We present a method for storing multiple models within a single set of parameters. Models can coexist in superposition and still be retrieved individually. In experiments with neural networks, we show that a surprisingly large number of models can be effectively stored within a single parameter instance. Furthermore, each of these models can undergo thousands of training steps without significantly interfering with other models within the superposition. This approach may be viewed as the online complement of compression: rather than reducing the size of a network after training, we make use of the unrealized capacity of a network during training.

Keywords: Deep Learning · Algorithms -> Representation Learning; Deep Learning -> Memory-Augmented Neural Networks; Neuroscience and Cognitive Science
[ Paper]  [ Slides] 

Author Information

Brian Cheung (UC Berkeley)
Alexander Terekhov (Awecom, Inc)
Yubei Chen (Berkeley AI Research UC Berkeley)
Pulkit Agrawal (UC Berkeley)
Bruno Olshausen (Redwood Center/UC Berkeley)

More from the Same Authors