Reading group on differentiable programming

This reading group takes place on Tuesdays from 2.15pm to 3.30pm, room 34_210 (to be confirmed) on a lax weekly basis. Its aim is to understand differentiable programming from a type-theoretic, Gallinette point-of-view.

Planning 2019-2010

• 01/10/19, 10h30, salle ABC: Damiano Mazza (LIPN, Université Paris 13), Backpropagation in the Simply-Typed Lambda-Calculus with Linear Negation
• 10/10/19, 13h30, salle 404 : Yann Regis Gianas, (IRIF, Unversité Paris Diderot)

Planning 2019:

• 22/01/19 : Marie Kerjean, Automatic Differentiation [1]
• 29/01/19 : Guilhem Jaber, Tensor flow [2] [3]
• 05/02/19 : Maxime Lucas [5] (salle du conseil 34_404)
• 12/02/19 :
• (19/02/19 : ? )
• 26/02/19 : Ambroise Lafont [12]
• 05/03/19 : Rémi Douence [10]
• 12/03/19 : Etienne Miquey [8]
• 19/03/19 : Pierre-Marie Pedrot [11] ( 13h45 !! )
• 26/03/19 : (séance remplacée par exposé de Xavier)
• 02/04/19 :
• 09/04/19 :
• (16/04/19 : ? )
• 23/04/19 : Marie [13]
• 30/04/19 : Pierre-Marie Pedrot, Dialectica and Linear Subsitution
• 07/05/19 :
• 14/05/19 : Jean-Simon Lemay, Introduction to Cartesian Differential Categories [18] [19]
• 16/05/19 : Mario Alvarez Picallo [7]
• 21/05/19 :
• 28/05/19 : Guilhem [16]
• 04/06/19 :
• ( 11/06/19 : TYPES )

References:

• [1] Barak Pearlmutter and Baydin, Automatic Differentiation of Algorithms for Machine Learning, or [1'] A. G. Baydin, B. A. Pearlmutter, A. A. Radul, and J. M. Siskind. Automatic differentiation in machine learning: a survey.
• [2] Tensorflow: a system for large-scale machine learning - Martin Abadi et al.
• [3] A Computational Model for TensorFlow (An Introduction) - Martin Abadi, Michael Isard and Derek G. Murray.
• [5] Conal Eliott, Compiling to categories
• [6] Conal Eliott, The simple Essence of Automatic Differentiation,
• [7] Thomas Ehrhard, Laurent Regnier, The Differential Lambda Calculus,
• [8] Conor McBride, The derivative of a regular type is its type of one-hole context
• [9] Michael Abbott, Thorsten Altenkirch, Neil Ghani, Conor McBride, ∂ for Data: Differentiating Data Structures
• [10] Fei Wang, James M. Decker, Xilun Wu, Grégory M. Essertel, Tiark Rompf, Backpropagation with Continuation Callbacks: Foundations for Efficient and Expressive Differentiable Programming
• [11] Barak A. Pearlmutter and Jeffrey Mark Siskind. Reverse-Mode AD in a functional framework: Lambda the ultimate backpropagator.
• [12] Thomas Beck and Herbert Fischer. The if-problem inautomatic differentiation.
• [13] Antonio Bucciarelli, Thomas Ehrhard, Giulio Manzonetto. Categorical Models for Simply Typed Resource Lambda-Calculus. 2010.
• [14] Some Principles of Differentiable Programming Languages, Invited talk by Plotkin at POPL 2018.
• [15] Efficient Differentiable Programming in a Functional Array-Processing Language. Amir Shaikhha, Andrew Fitzgibbon, Dimitrios Vytiniotis, Simon Peyton Jones, Christoph Koch. 2018
• [16] A theory of changes for higher-order languages: incrementalizing λ-calculi by static differentiation. Yufei Cai, Paolo G. Giarrusso, , Tillmann Rendel, Klaus Ostermann. 2013
• [17] Change Actions: Models of Generalised Differentiation Mario Alvarez-Picallo, C.-H. Luke Ong. 2019
• [18] Cartesian Differential Categories
• [19] Cartesian Closed Differential Categories