Timezone: »

 
Poster
Tangent: Automatic differentiation using source-code transformation for dynamically typed array programming
Bart van Merriënboer · Dan Moldovan · Alexander Wiltschko

Wed Dec 05 07:45 AM -- 09:45 AM (PST) @ Room 210 #45
in Wed Poster Session A »

The need to efficiently calculate first- and higher-order derivatives of increasingly complex models expressed in Python has stressed or exceeded the capabilities of available tools. In this work, we explore techniques from the field of automatic differentiation (AD) that can give researchers expressive power, performance and strong usability. These include source-code transformation (SCT), flexible gradient surgery, efficient in-place array operations, and higher-order derivatives. We implement and demonstrate these ideas in the Tangent software library for Python, the first AD framework for a dynamic language that uses SCT.

Keywords: Hardware and Systems · Software Toolkits
[ Paper] 

Author Information

Bart van Merriënboer (MILA, Google)
Dan Moldovan (Google)
Alexander Wiltschko (Google Brain)

More from the Same Authors