I have successfully defended my thesis in December 2016.
I am going to start as Machine Learning Research Scientist at Apple starting from July 2017.

PhD student
Member LEAR Team, INRIA Rhone-Alpes, Grenoble, France

INRIA Rhone-Alpes 655 Avenue de l'Europe 38330 Montbonnot France

Phone Office E-mail

: : :

+33 4 76 61 52 25 H117 shreyas.saxena@inria.fr

Short Bio

I am in the third and final year of my Ph.D at INRIA. I am pursuing my PhD under the supervision of Jakob Verbeek and Cordelia Schmid . I received my Master's degree in Computer Science from the Université Joseph Fourier (UJF), Grenoble, France and carried out my thesis under Jakob Verbeek . My thesis was on Metric learning for face verification.

You can find my CV here.
I can be reached by email at shreyas[dot]saxena2[at]gmail[dot]com.

---

I am working on data and method driven machine learning models for the image/scene understanding, image segmentation, verification and retrieval and dimensionality reduction methods. At the moment, following topics are of interest:

• Architecture learning.
• Learning dynamic CNN models, where the network parameters are conditioned on the input.
• Training extremely deep networks via teacher student paradigm.

Topics I have not worked on directly, but keen on exploring in future:

• Generative adversarial networks for image synthesis from text, and sampling from the latent space.
• Memory networks in the context of question and answering.

Recent Work

---

	Convolutional Neural Fabrics Shreyas Saxena, Jakob Verbeek, Neural Information Processing Systems (NIPS), Dec 2016, Barcelona, Spain.

	Heterogeneous Face Recognition with CNNs Shreyas Saxena, Jakob Verbeek, ECCV TASK-CV workshop, Oct 2016, Amsterdam, Netherlands.

	Coordinated Local Metric Learning Shreyas Saxena, Jakob Verbeek, ICCV ChaLearn Looking at People workshop, Dec 2015, Santiago, Chile.