no code implementations • ICLR 2018 • David Macêdo, Cleber Zanchettin, Adriano L. I. Oliveira, Teresa Ludermir
Besides, statistical significant performance assessments (p<0. 05) showed DReLU enhanced the test accuracy presented by ReLU in all scenarios.
no code implementations • ICLR 2018 • David Macêdo, Cleber Zanchettin, Teresa Ludermir
The quality of the features used in visual recognition is of fundamental importance for the overall system.
1 code implementation • 15 Aug 2019 • David Macêdo, Tsang Ing Ren, Cleber Zanchettin, Adriano L. I. Oliveira, Teresa Ludermir
Consequently, we propose IsoMax, a loss that is isotropic (distance-based) and produces high entropy (low confidence) posterior probability distributions despite still relying on cross-entropy minimization.
2 code implementations • 7 Jun 2020 • David Macêdo, Tsang Ing Ren, Cleber Zanchettin, Adriano L. I. Oliveira, Teresa Ludermir
In this paper, we argue that the unsatisfactory out-of-distribution (OOD) detection performance of neural networks is mainly due to the SoftMax loss anisotropy and propensity to produce low entropy probability distributions in disagreement with the principle of maximum entropy.
no code implementations • 13 Oct 2020 • Felipe Farias, Teresa Ludermir, Carmelo Bastos-Filho
We propose a Similarity-Based Stratified Splitting (SBSS) technique, which uses both the output and input space information to split the data.
no code implementations • 17 Feb 2021 • David Macêdo, Pedro Dreyer, Teresa Ludermir, Cleber Zanchettin
We compared the proposed approach with commonly used adaptive learning rate schedules such as Adam, RMSProp, and Adagrad.
no code implementations • NeurIPS 2021 • David Macêdo, Teresa Ludermir
The entropic out-of-distribution detection solution comprises the IsoMax loss for training and the entropic score for out-of-distribution detection.
1 code implementation • 30 May 2021 • David Macêdo, Teresa Ludermir
The IsoMax loss works as a drop-in replacement of the SoftMax loss (i. e., the combination of the output linear layer, the SoftMax activation, and the cross-entropy loss) because swapping the SoftMax loss with the IsoMax loss requires no changes in the model's architecture or training procedures/hyperparameters.
no code implementations • 5 Jul 2021 • Danielle Silva, Teresa Ludermir
Training of Artificial Neural Networks is a complex task of great importance in supervised learning problems.
no code implementations • 10 Jul 2021 • Tarsicio Lucas, Teresa Ludermir, Ricardo Prudencio, Carlos Soares
In the current work, we performed a case study using meta-learning to choose the number of hidden nodes for MLP networks, which is an important parameter to be defined aiming a good networks performance.
no code implementations • 18 Jul 2021 • Anderson da Silva, Teresa Ludermir
This works proposes a methodology to searching for automatically Artificial Neural Networks (ANN) by using Cellular Genetic Algorithm (CGA).
no code implementations • 29 Jul 2021 • Gecynalda Gomes, Teresa Ludermir
A methodology for the global optimization of this family of activation functions with free parameter and the weights of the connections between the processing units of the neural network is used.
no code implementations • 2 Aug 2021 • Jefferson Souza, Teresa Ludermir
This work investigates the use of two-stage clustering methods.
1 code implementation • 12 May 2022 • David Macêdo, Cleber Zanchettin, Teresa Ludermir
In this paper, we propose training deterministic neural networks using our DisMax loss, which works as a drop-in replacement for the usual SoftMax loss (i. e., the combination of the linear output layer, the SoftMax activation, and the cross-entropy loss).