Rue Simone Iff 2, Voie DQ12

75012 Paris, France

Phone: +33 1 80 49 43 76

Office: C413 (level 4)

E-mail: adrien.taylor [at] inria.fr

I am currently a researcher at Inria, within the SIERRA team (led by Francis Bach). I was a postdoctoral researcher in the same team in 2017-2019. Prior to that, I was at Université Catholique
de Louvain, in the department of mathematical engineering (part of the ICTEAM institute), where I was holding a F.R.S.-FNRS FRIA scholarship for my PhD under the supervision of François Glineur and Julien Hendrickx.

Broad research interests; current focus on theory and algorithms for convex optimization and monotone inclusion problems.

**Thesis - ** My thesis (under the supervision of François Glineur and Julien Hendrickx) had the chance to be awarded the ICTEAM thesis award for 2018, the IBM-FNRS innovation award for 2018, and to be a finalist for the AW Tucker prize for 2018. In addition, we received a 2017 best paper award in Optimization Letters, for a joint work with Etienne de Klerk and François Glineur (for this paper).

**Software - ** The current version of the **Performance EStimation TOolbox (PESTO)** is available from here (full manual here). The numerical worst-case analyses from PEP can now be performed just by writting the algorithms just as you would implement them. This work will be presented at CDC2017 in Melbourne; the proceeding offers a nice introduction to the topic of automated worst-case analyses using performance estimation problems.

**Tutorial - ** If you want to get familiar with performance estimation problems, you might be interested in checking this blog post. The performance estimation toolbox (github) is probably also a good way to try the approach without spending too much time in the details, and the user guide provides the list of examples that are contained within the toolbox.

- Introduction to computer-assisted proofs in optimization and numerical analysis (upcoming)
- Computer-aided analyses for first-order methods (via semidefinite programming) (Summer school on convex optimization, slides, video)
- Computer-aided analyses of first-order methods (via semidefinite programming) [Very Long version] (CCIMI seminar Cambridge, slides)
- Non-asymptotic and computer-aided analyses via potential functions (COLT 2019, slides, video)
- Performance estimation toolbox (PESTO) (upcoming)
- Efficient first-order methods for convex minimization: a constructive approach (upcoming)
- Computer-aided worst-case analyses for operator splitting (ICCOPT 2019, slides)

- (2020) E.K. Ryu, A.B. Taylor, C. Bergeling, P. Giselsson. Operator Splitting Performance Estimation: Tight contraction factors and optimal parameter selection. SIAM Journal on Optimization (to appear, arXiv, codes)
- (2020) E. de Klerk, F. Glineur, A.B Taylor. Worst-case convergence analysis of inexact gradient and Newton methods through semidefinite programming performance estimation. SIAM Journal on Optimization (to appear, arXiv, codes)
- (2019) Yoel Drori, A.B. Taylor. Efficient first-order methods for convex minimization: a constructive approach. Mathematical Programming. (arXiv, Springer, codes)
- (2018) A.B. Taylor, J.M. Hendrickx, F. Glineur. Exact worst-case convergence rates of the proximal gradient method for composite convex minimization. Journal of Optimization Theory and Algorithms. (arXiv, Springer, codes)
- (2017) A.B. Taylor, J.M. Hendrickx, F. Glineur. Smooth strongly convex interpolation and exact worst-case performance of first-order methods. Mathematical Programming 161 (1-2), 307–345. (arXiv, Springer, codes)
- (2017) A.B. Taylor, J.M. Hendrickx, F. Glineur. Exact worst-case performance of first-order methods for composite convex optimization. SIAM Journal on Optimization 27(3), 1283–1313. (arXiv, SIAM, codes)
- (2017,
**Best paper award**) E. de Klerk, F. Glineur, A.B. Taylor. On the worst-case complexity of the gradient method with exact line search for smooth strongly convex functions. Optimization Letters 11(7), 1185–1199. (Springer)

- (2020) M. Barré, A.B. Taylor, A. d'Aspremont. Complexity Guarantees for Polyak Steps with Momentum. Conference on Learning Theory (COLT). (arXiv, codes)
- (2019) A.B. Taylor, F. Bach. Stochastic first-order methods: non-asymptotic and computer-aided analyses via potential functions. Conference on Learning Theory (COLT). (PMLR, arXiv, codes)
- (2018) A.B. Taylor, B. Van Scoy, L. Lessard. Lyapunov Functions for First-Order Methods: Tight Automated Convergence Guarantees; Proceedings of the 35th International Conference on Machine Learning (ICML), PMLR 80:4904-4913, 2018. (PMRL, arxiv, codes)
- (2017) A.B. Taylor, J.M. Hendrickx, F. Glineur. Performance Estimation Toolbox (PESTO): automated worst-case analysis of first-order optimization methods. Proceedings of the 56th IEEE Conference on Decision and Control (CDC). (IEEE, local, toolbox)

- (2017,
**Performance EStimation TOolbox (PESTO)**) The current version of the Performance EStimation TOolbox (PESTO) is available from here. The numerical worst-case analyses from PEP can now be performed just by writting the algorithms just as you would implement them. - Want to reproduce the results of one of those previous works ? Check the codes referenced above or go to my github profile!

- Expected Improvement criteria for multi-objective optimization, Master thesis, June 2011. (Advisers: F. Glineur and M. Diehl)
- Convex Interpolation and Performance Estimation of First-order Methods for Convex Optimization, PhD Thesis, January 2017. (Advisers: F. Glineur and J. Hendrickx)

- Numerical Methods (LFSAB 1104, with Vincent Legat), Monitoring student, 2008.
- Automatic Control: Theory and Applications (LINMA2671, with Julien Hendrickx), TA, 2012, 2013, 2014.
- System Idenfitication (LINMA2875, with Julien Hendrickx), TA, 2013, 2014, 2015, 2016.
- Mathematics 2 (LFSAB1102, with Enrico Vitale, François Glineur and Roland Keunings), TA, 2013.
- Economy (LFSAB1803, with Julien Hendrickx), TA, 2013.
- Dynamical Systems Modeling and Analysis (LINMA2370, with Jean-Charles Delvenne and Denis Dochain), TA, 2013.
- Project in Applied Mathematics (LFSAB1507, with Julien Hendrickx, François Glineur, Pierre-Antoine Absil and Yurii Nesterov), TA, 2014, 2015, 2016.
- Numerical Analysis (LINMA1170, with Paul Van Dooren), TA, 2014.
- Matrix Theory (LINMA2380, with Paul Van Dooren), TA, 2016.

At UCLouvain, I used to share my office with great colleagues: François Gonze and Pierre-Yves Chevalier.