daouda sow - Academia.edu (original) (raw)

Papers by daouda sow

Research paper thumbnail of A Constrained Optimization Approach to Bilevel Optimization with Multiple Inner Minima

ArXiv, 2022

Bilevel optimization has found extensive applications in modern machine learning problems such as... more Bilevel optimization has found extensive applications in modern machine learning problems such as hyperparameter optimization, neural architecture search, meta-learning, etc. While bilevel problems with a unique inner minimal point (e.g., where the inner function is strongly convex) are well understood, bilevel problems with multiple inner minimal points remains to be a challenging and open problem. Existing algorithms designed for such a problem were applicable to restricted situations and do not come with the full guarantee of convergence. In this paper, we propose a new approach, which convert the bilevel problem to an equivalent constrained optimization, and then the primal-dual algorithm can be used to solve the problem. Such an approach enjoys a few advantages including (a) addresses the multiple inner minima challenge; (b) features fully first-order efficiency without involving second-order Hessian and Jacobian computations, as opposed to most existing gradient-based bilevel ...

Research paper thumbnail of Idiopathic Juxtafoveolar Retinal Telangiectasis

The Official Scientific Journal of Delhi Ophthalmological Society, May 27, 2011

Research paper thumbnail of A Novel Hessian-Free Bilevel Optimizer via Evolution Strategies

Bilevel optimization has arisen as a powerful tool for solving many modern machine learning probl... more Bilevel optimization has arisen as a powerful tool for solving many modern machine learning problems. However, due to the nested structure of bilevel optimization, even gradient-based methods require secondorder derivative approximations via Jacobianor/and Hessian-vector computations, which can be very costly in practice. In this work, we propose a novel Hessian-free bilevel algorithm, which adopts the Evolution Strategies (ES) method to approximate the response Jacobian matrix in the hypergradient of the bilevel problem, and hence fully eliminates all second-order computations. We call our algorithm as ESJ (which stands for the ES-based Jacobian method) and further extend it to the stochastic setting as ESJ-S. Theoretically, we show that both ESJ and ESJ-S are guaranteed to converge. Experimentally, we demonstrate that the proposed algorithms outperform baseline bilevel optimizers on various bilevel problems. Particularly, in our experiment on few-shot meta-learning of ResNet-12 ne...

Research paper thumbnail of A Sequential Guiding Network with Attention for Image Captioning

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019

The recent advances of deep learning in both computer vision (CV) and natural language processing... more The recent advances of deep learning in both computer vision (CV) and natural language processing (NLP) provide us a new way of understanding semantics, by which we can deal with more challenging tasks such as automatic description generation from natural images. In this challenge, the encoder-decoder framework has achieved promising performance when a convolutional neural network (CNN) is used as image encoder and a recurrent neural network (RNN) as decoder. In this paper, we introduce a sequential guiding network that guides the decoder during word generation. The new model is an extension of the encoder-decoder framework with attention that has an additional guiding long short-term memory (LSTM) and can be trained in an end-to-end manner by using image/descriptions pairs. We validate our approach by conducting extensive experiments on a benchmark dataset, i.e., MS COCO Captions. The proposed model achieves significant improvement comparing to the other state-ofthe-art deep learning models.

Research paper thumbnail of Three-Step Charging Algorithm for Lead-Acid Batteries

This work presents the development of a solar regulator which manages the charge and discharge of... more This work presents the development of a solar regulator which manages the charge and discharge of a (lead) battery installed in a photovoltaic system in order to extend its lifetime. The regulator is controlled by a microcontroller (PIC16F877A) and protects the battery against overcharging, deep discharge, but also against temperature drifts. The operating principle is based on the control of a DC-DC converter by a rectangular signal MLI generated by the microcontroller. In addition to the protection function of the regulator, there is included a control and monitoring panel consisting of a visualization interface on which the system quantities can be observed. Thus, it will be given to the user to be able to act on the system. This display interface uses as a display an LCD screen and LEDs. Simulation results are presented to illustrate the operation of the proposed solar controller.

Research paper thumbnail of ES-Based Jacobian Enables Faster Bilevel Optimization

ArXiv, 2021

Bilevel optimization (BO) has arisen as a powerful tool for solving many modern machine learning ... more Bilevel optimization (BO) has arisen as a powerful tool for solving many modern machine learning problems. However, due to the nested structure of BO, existing gradient-based methods require secondorder derivative approximations via Jacobianor/and Hessian-vector computations, which can be very costly in practice, especially with large neural network models. In this work, we propose a novel BO algorithm, which adopts Evolution Strategies (ES) based method to approximate the response Jacobian matrix in the hypergradient of BO, and hence fully eliminates all second-order computations. We call our algorithm as ESJ (which stands for the ES-based Jacobian method) and further extend it to the stochastic setting as ESJ-S. Theoretically, we characterize the convergence guarantee and computational complexity for our algorithms. Experimentally, we demonstrate the superiority of our proposed algorithms compared to the state of the art methods on various bilevel problems. Particularly, in our ex...

Research paper thumbnail of Development of a Solar Controller with MLI Control

Circuits and Systems, 2018

This work presents the development of a solar regulator which manages the charge and discharge of... more This work presents the development of a solar regulator which manages the charge and discharge of a (lead) battery installed in a photovoltaic system in order to extend its lifetime. The regulator is controlled by a microcontroller (PIC16F877A) and protects the battery against overcharging, deep discharge, but also against temperature drifts. The operating principle is based on the control of a DC-DC converter by a rectangular signal MLI generated by the microcontroller. In addition to the protection function of the regulator, there is included a control and monitoring panel consisting of a visualization interface on which the system quantities can be observed. Thus, it will be given to the user to be able to act on the system. This display interface uses as a display an LCD screen and LEDs. Simulation results are presented to illustrate the operation of the proposed solar controller.

Research paper thumbnail of Immaginario collettivo e identità locale. La valorizzazione turistica del patrimonio culturale fra Tunisia e Sardegna

ABSTRACT Il volume mostra come nel Mediterraneo il turismo culturale sia fortemente dipendente da... more ABSTRACT Il volume mostra come nel Mediterraneo il turismo culturale sia fortemente dipendente dal turismo costiero. Il testo illustra le dinamiche del processo di valorizzazione istituzionale, in particolare in Tunisia, e il peso dell’UNESCO nelle dinamiche territoriali locali. Sono analizzate le ricadute territoriali potenziali della creazione di un parco archeologico in aree in ritardo di sviluppo, e individuate una serie di pratiche in grado di favorire la cooperazione fra le due sponde del Mediterraneo.

Research paper thumbnail of A Constrained Optimization Approach to Bilevel Optimization with Multiple Inner Minima

ArXiv, 2022

Bilevel optimization has found extensive applications in modern machine learning problems such as... more Bilevel optimization has found extensive applications in modern machine learning problems such as hyperparameter optimization, neural architecture search, meta-learning, etc. While bilevel problems with a unique inner minimal point (e.g., where the inner function is strongly convex) are well understood, bilevel problems with multiple inner minimal points remains to be a challenging and open problem. Existing algorithms designed for such a problem were applicable to restricted situations and do not come with the full guarantee of convergence. In this paper, we propose a new approach, which convert the bilevel problem to an equivalent constrained optimization, and then the primal-dual algorithm can be used to solve the problem. Such an approach enjoys a few advantages including (a) addresses the multiple inner minima challenge; (b) features fully first-order efficiency without involving second-order Hessian and Jacobian computations, as opposed to most existing gradient-based bilevel ...

Research paper thumbnail of Idiopathic Juxtafoveolar Retinal Telangiectasis

The Official Scientific Journal of Delhi Ophthalmological Society, May 27, 2011

Research paper thumbnail of A Novel Hessian-Free Bilevel Optimizer via Evolution Strategies

Bilevel optimization has arisen as a powerful tool for solving many modern machine learning probl... more Bilevel optimization has arisen as a powerful tool for solving many modern machine learning problems. However, due to the nested structure of bilevel optimization, even gradient-based methods require secondorder derivative approximations via Jacobianor/and Hessian-vector computations, which can be very costly in practice. In this work, we propose a novel Hessian-free bilevel algorithm, which adopts the Evolution Strategies (ES) method to approximate the response Jacobian matrix in the hypergradient of the bilevel problem, and hence fully eliminates all second-order computations. We call our algorithm as ESJ (which stands for the ES-based Jacobian method) and further extend it to the stochastic setting as ESJ-S. Theoretically, we show that both ESJ and ESJ-S are guaranteed to converge. Experimentally, we demonstrate that the proposed algorithms outperform baseline bilevel optimizers on various bilevel problems. Particularly, in our experiment on few-shot meta-learning of ResNet-12 ne...

Research paper thumbnail of A Sequential Guiding Network with Attention for Image Captioning

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019

The recent advances of deep learning in both computer vision (CV) and natural language processing... more The recent advances of deep learning in both computer vision (CV) and natural language processing (NLP) provide us a new way of understanding semantics, by which we can deal with more challenging tasks such as automatic description generation from natural images. In this challenge, the encoder-decoder framework has achieved promising performance when a convolutional neural network (CNN) is used as image encoder and a recurrent neural network (RNN) as decoder. In this paper, we introduce a sequential guiding network that guides the decoder during word generation. The new model is an extension of the encoder-decoder framework with attention that has an additional guiding long short-term memory (LSTM) and can be trained in an end-to-end manner by using image/descriptions pairs. We validate our approach by conducting extensive experiments on a benchmark dataset, i.e., MS COCO Captions. The proposed model achieves significant improvement comparing to the other state-ofthe-art deep learning models.

Research paper thumbnail of Three-Step Charging Algorithm for Lead-Acid Batteries

This work presents the development of a solar regulator which manages the charge and discharge of... more This work presents the development of a solar regulator which manages the charge and discharge of a (lead) battery installed in a photovoltaic system in order to extend its lifetime. The regulator is controlled by a microcontroller (PIC16F877A) and protects the battery against overcharging, deep discharge, but also against temperature drifts. The operating principle is based on the control of a DC-DC converter by a rectangular signal MLI generated by the microcontroller. In addition to the protection function of the regulator, there is included a control and monitoring panel consisting of a visualization interface on which the system quantities can be observed. Thus, it will be given to the user to be able to act on the system. This display interface uses as a display an LCD screen and LEDs. Simulation results are presented to illustrate the operation of the proposed solar controller.

Research paper thumbnail of ES-Based Jacobian Enables Faster Bilevel Optimization

ArXiv, 2021

Bilevel optimization (BO) has arisen as a powerful tool for solving many modern machine learning ... more Bilevel optimization (BO) has arisen as a powerful tool for solving many modern machine learning problems. However, due to the nested structure of BO, existing gradient-based methods require secondorder derivative approximations via Jacobianor/and Hessian-vector computations, which can be very costly in practice, especially with large neural network models. In this work, we propose a novel BO algorithm, which adopts Evolution Strategies (ES) based method to approximate the response Jacobian matrix in the hypergradient of BO, and hence fully eliminates all second-order computations. We call our algorithm as ESJ (which stands for the ES-based Jacobian method) and further extend it to the stochastic setting as ESJ-S. Theoretically, we characterize the convergence guarantee and computational complexity for our algorithms. Experimentally, we demonstrate the superiority of our proposed algorithms compared to the state of the art methods on various bilevel problems. Particularly, in our ex...

Research paper thumbnail of Development of a Solar Controller with MLI Control

Circuits and Systems, 2018

This work presents the development of a solar regulator which manages the charge and discharge of... more This work presents the development of a solar regulator which manages the charge and discharge of a (lead) battery installed in a photovoltaic system in order to extend its lifetime. The regulator is controlled by a microcontroller (PIC16F877A) and protects the battery against overcharging, deep discharge, but also against temperature drifts. The operating principle is based on the control of a DC-DC converter by a rectangular signal MLI generated by the microcontroller. In addition to the protection function of the regulator, there is included a control and monitoring panel consisting of a visualization interface on which the system quantities can be observed. Thus, it will be given to the user to be able to act on the system. This display interface uses as a display an LCD screen and LEDs. Simulation results are presented to illustrate the operation of the proposed solar controller.

Research paper thumbnail of Immaginario collettivo e identità locale. La valorizzazione turistica del patrimonio culturale fra Tunisia e Sardegna

ABSTRACT Il volume mostra come nel Mediterraneo il turismo culturale sia fortemente dipendente da... more ABSTRACT Il volume mostra come nel Mediterraneo il turismo culturale sia fortemente dipendente dal turismo costiero. Il testo illustra le dinamiche del processo di valorizzazione istituzionale, in particolare in Tunisia, e il peso dell’UNESCO nelle dinamiche territoriali locali. Sono analizzate le ricadute territoriali potenziali della creazione di un parco archeologico in aree in ritardo di sviluppo, e individuate una serie di pratiche in grado di favorire la cooperazione fra le due sponde del Mediterraneo.