Luis Argerich

Moacir Antonelli Ponti, Lucas de Angelis Oliveira, Mathias Esteban et al.

Real world datasets contain incorrectly labeled instances that hamper the performance of the model and, in particular, the ability to generalize out of distribution. Also, each example might have different contribution towards learning. This motivates studies to better understanding of the role of data instances with respect to their contribution in good metrics in models. In this paper we propose a method based on metrics computed from training dynamics of Gradient Boosting Decision Trees (GBDTs) to assess the behavior of each training example. We focus on datasets containing mostly tabular or structured data, for which the use of Decision Trees ensembles are still the state-of-the-art in terms of performance. Our methods achieved the best results overall when compared with confident learning, direct heuristics and a robust boosting algorithm. We show results on detecting noisy labels in order clean datasets, improving models' metrics in synthetic and real public datasets, as well as on a industry case in which we deployed a model based on the proposed solution.

Emanuele Luzio, Moacir Antonelli Ponti, Christian Ramirez Arevalo et al.

Machine learning models typically focus on specific targets like creating classifiers, often based on known population feature distributions in a business context. However, models calculating individual features adapt over time to improve precision, introducing the concept of decoupling: shifting from point evaluation to data distribution. We use calibration strategies as strategy for decoupling machine learning (ML) classifiers from score-based actions within business logic frameworks. To evaluate these strategies, we perform a comparative analysis using a real-world business scenario and multiple ML models. Our findings highlight the trade-offs and performance implications of the approach, offering valuable insights for practitioners seeking to optimize their decoupling efforts. In particular, the Isotonic and Beta calibration methods stand out for scenarios in which there is shift between training and testing data.

Leandro Palma, Luis Argerich

In this work we present a modification in the conventional flow of information through a LSTM network, which we consider well suited for RNNs in general. The modification leads to a iterative scheme where the computations performed by the LSTM cell are repeated over a constant input and cell state values, while updating the hidden state a finite number of times. We provide theoretical and empirical evidence to support the augmented capabilities of the iterative scheme and show examples related to language modeling. The modification yields an enhancement in the model performance comparable with the original model augmented more than 3 times in terms of the total amount of parameters.

Juan Andrés Laura, Gabriel Masi, Luis Argerich

In recent studies [1][13][12] Recurrent Neural Networks were used for generative processes and their surprising performance can be explained by their ability to create good predictions. In addition, data compression is also based on predictions. What the problem comes down to is whether a data compressor could be used to perform as well as recurrent neural networks in natural language processing tasks. If this is possible,then the problem comes down to determining if a compression algorithm is even more intelligent than a neural network in specific tasks related to human language. In our journey we discovered what we think is the fundamental difference between a Data Compression Algorithm and a Recurrent Neural Network.

Luis Argerich, Natalia Golmar

In this paper we propose the creation of generic LSH families for the angular distance based on Johnson-Lindenstrauss projections. We show that feature hashing is a valid J-L projection and propose two new LSH families based on feature hashing. These new LSH families are tested on both synthetic and real datasets with very good results and a considerable performance improvement over other LSH families. While the theoretical analysis is done for the angular distance, these families can also be used in practice for the euclidean distance with excellent results [2]. Our tests using real datasets show that the proposed LSH functions work well for the euclidean distance.

Luis Argerich, Joaquín Torré Zaffaroni, Matías J Cano

In this paper we propose the application of feature hashing to create word embeddings for natural language processing. Feature hashing has been used successfully to create document vectors in related tasks like document classification. In this work we show that feature hashing can be applied to obtain word embeddings in linear time with the size of the data. The results show that this algorithm, that does not need training, is able to capture the semantic meaning of words. We compare the results against GloVe showing that they are similar. As far as we know this is the first application of feature hashing to the word embeddings problem and the results indicate this is a scalable technique with practical results for NLP applications.

Federico Barrios, Federico López, Luis Argerich et al.

This article presents new alternatives to the similarity function for the TextRank algorithm for automatic summarization of texts. We describe the generalities of the algorithm and the different functions we propose. Some of these variants achieve a significative improvement using the same metrics and dataset as the original publication.