#### Filter Results:

- Full text PDF available (9)

#### Publication Year

2009

2018

- This year (5)
- Last 5 years (18)
- Last 10 years (20)

#### Publication Type

#### Co-author

#### Journals and Conferences

Learn More

- Pratik Chaudhari, Anna Choromanska, +6 authors Riccardo Zecchina
- ArXiv
- 2016

This paper proposes a new optimization algorithm called Entropy-SGD for training deep neural networks that is motivated by the local geometry of the energy landscape. Local extrema with lowâ€¦ (More)

Visual representations are defined in terms of minimal sufficient statistics of visual data, for a class of tasks, that are also invariant to nuisance variability. Minimal sufficiency guarantees thatâ€¦ (More)

We study a theoretical model that connects deep learning to finding the ground state of the Hamiltonian of a spherical spin glass. Existing results motivated from statistical physics show that deepâ€¦ (More)

- Pratik Chaudhari, Stefano Soatto
- ArXiv
- 2017

Stochastic gradient descent (SGD) is widely believed to perform implicit regularization when used to train deep neural networks, but the precise manner in which this occurs has thus far been elusive.â€¦ (More)

- Pratik Chaudhari, Stefano Soatto
- ArXiv
- 2015

- Luis I. Reyes Castro, Pratik Chaudhari, Jana Tumova, Sertac Karaman, Emilio Frazzoli, Daniela Rus
- 52nd IEEE Conference on Decision and Control
- 2013

This paper studies the problem of control strategy synthesis for dynamical systems with differential constraints to fulfill a given reachability goal while satisfying a set of safety rules.â€¦ (More)

We establish connections between non-convex optimization methods for training deep neural networks (DNNs) and the theory of partial differential equations (PDEs). In particular, we focus onâ€¦ (More)

- Valerio Varricchio, Pratik Chaudhari, Emilio Frazzoli
- 2014 IEEE International Conference on Roboticsâ€¦
- 2014

This paper investigates motion-planning using formal language specifications for dynamical systems with differential constraints. In particular, we focus on process algebra as a language to specifyâ€¦ (More)

We propose a new algorithm called Parle for parallel training of deep networks that converges 2-4Ã— faster than a data-parallel implementation of SGD, while achieving significantly improved errorâ€¦ (More)

- Pratik Chaudhari, Sertac Karaman, David Hsu, Emilio Frazzoli
- 2013 American Control Conference
- 2013

This paper focuses on a continuous-time, continuous-space formulation of the stochastic optimal control problem with nonlinear dynamics and observation noise. We lay the mathematical foundations toâ€¦ (More)