Brazil

This work presents a study on an ethical issue in Artificial Intelligence related to the presence of racist biases by detecting faces in images. Our analyses were performed on a real-world system designed to detect fraud in public transportation in Salvador (Brazil). Our experiments were conducted by taking into account three steps. Firstly, we individually analyzed a sample of images and added specific labels related to the users&#39; gender and race. Then, we used well-defined detectors, based on different Convolutional Neural Network architectures, to find faces in the previously labeled images. Finally, we used statistical tests to assess whether or not there is some relation between the error rates and such labels. According to our results, we had noticed important biases, thus leading to higher error rates when images were taken from black people. We also noticed errors are more likely in both black men and women. Based on our conclusions, we recall the risk of deploying computational software that might affect minority groups that are historically neglected.

BRACIS 2021

Ethics of AI: Do the face detection models act with prejudice?

ethic \and deep learning

racism

face detection

Join AAPT and Physics Educators worldwide in developing and strengthening your professional connections by attending the 2021 AAPT Virtual Summer Meeting. AAPT meetings are a great way to reflect on your teaching, recommit to best practices and re-energize your career with new research, products, and techniques.

Listen to internationally known plenary speakers, attend sessions, workshops, and special activities. Attending the virtual summer meeting is an excellent opportunity to share your expertise and learn from and network with leaders in the field.

Please log in with your AAPT account or register for the event if you haven't already.

If you are registered please note that you need to be logged in with your AAPT email / credentials in order to get access.

AAPT SM 21

This work presents a study on an ethical issue in Artificial Intelligence related to the presence of racist biases by detecting faces in images. Our analyses were performed on a real-world system designed to detect fraud in public transportation in Salvador (Brazil). Our experiments were conducted by taking into account three steps. Firstly, we individually analyzed a sample of images and added specific labels related to the users' gender and race. Then, we used well-defined detectors, based on different Convolutional Neural Network architectures, to find faces in the previously labeled images. Finally, we used statistical tests to assess whether or not there is some relation between the error rates and such labels. According to our results, we had noticed important biases, thus leading to higher error rates when images were taken from black people. We also noticed errors are more likely in both black men and women. Based on our conclusions, we recall the risk of deploying computational software that might affect minority groups that are historically neglected.

technical paper

**BRACIS** originated from the combination of the two most important scientific events in Brazil in Artificial Intelligence (AI) and Computational Intelligence (CI): the Brazilian Symposium on Artificial Intelligence – **SBIA** (22 editions), and the Brazilian Symposium on Neural Networks – **SBRN** (13 editions). **ENIAC** is a national forum for researchers, practitioners, educators and students to present and discuss innovations, trends, experiences, developments and work in progress in the fields of Artificial and Computational Intelligence. **STIL** is the bi-annual Language Technology event.
All events are supported by the Brazilian Computer Society – Special Interest Groups: **CE-IA**, **CE-IC** and **CE-PLN**.

To access the 10th Brazilian Conference on Intelligent Systems (**BRACIS**), you need to register for the conference.  Please visit the Conference official site to register

[Register now](http://c4ai.inova.usp.br/bracis2021/)

Register now

Organized by C4AI, the 10th Brazilian Conference on Intelligent Systems (BRACIS) will take place from November 29th to December 3rd, 2021.

We propose in this paper a framework of prioritised logic programs (PLP) to represent priority information explicitly in a program. Differently of others approaches, we do not restrain the preference relation only to literals, but to sets of literals. As consequence, we can express in PLPs sophisticated forms of preferences without changing the programs or introducing new atoms to obtain artificially the intended preferences. Besides, inspired by various developments in the literature on preference, we present a comprehensive and systematic treatment to deal with preferences in logic programming. In fact, we introduced 32 different criteria (semantics) to establish preference between partial stable models as well as those semantics whose definition depends on partial stable models. We show some properties of our framework; in particular, we guarantee these semantics for PLP generalise their counterparts for logic programs without preferences.

A Systematic Approach to Define Semantics for Prioritised Logic Programs

Value-based health care management models require a precise accounting of health indexes such as risk events monitoring, clinical conditions, patient handling and outcomes. Currently this accounting is performed by manually reading and searching through electronic health records for these indexes. Our research proposes a way to make this an autonomous task that is performed by a computer using a portuguese free-text concept classifier model based on ontologies. To validate our model we tested it on digital clinical records from 191 patients under ischemic stroke care. We have selected 30 management indexes to be identified in these texts. Our model reached, on average 56,8 % of f1-score, varying from 5,83 % to 94,78 % f1-score across different management indexes.

Ontology based classification of electronic health records to support value-based health care

Tactical asset allocation is an essential method for defining a profitable portfolio for a given period. An analyst usually creates a tactical asset portfolio through technical analysis, a process subjective to the analyst's knowledge and interpretations. Another aspect that can directly influence the quality of a portfolio is the number of assets considered for analysis. Human analysts tend to focus on a pre-defined group of assets, limiting choices, and, consequently, the possibility of better results. This work proposes the Stock Network Portfolio Allocation (SNPA) algorithm for the automatic recommendation of a stock portfolio, aiming to maximize profit and minimize risk. The proposed method considers a possibly large set of assets represented as a complex network. In which the nodes represent assets, and the edges stand for the correlation between their returns. Portfolio allocation is done through a random walk on the stock network, selecting, in the end, the most visited nodes (stocks). We conducted investment simulations on Brazilian stocks from the IBrX100 index, for 24 month periods, from Jan. 2018 to Dec. 2019. We compare the results with portfolio strategies: Ibovespa index (IBOV), classic Markowitz’s mean-variance portfolio (MV), Mean Absolute Deviation (MAD) portfolio, Conditional Value at Risk (CVaR), and Hierarchical Risk Parity (HRP). The Shape Ratio (SR), Maximum Drawdown (MDD), and Cumulated Wealth (CW) were used as performance metrics. The SNPA algorithm demonstrated its effectiveness, presenting a CW of 236.3%, being 203.5% above MV portfolio; 181.7% above CVaR portfolio; 175.6% above MAD portfolio, 184.6% above IBOV index, and 165.1% above HRP. SNPA also surpassed the benchmarks considering the performance metrics SR and MDD, with values 0.67 and -1.37 respectively, the best results among the benchmarks were observed by the HRP strategy, with 0.48 in SR index, and MV with -1.39 in MDD index.

Tactical Asset Allocation through Random Walk on Stock Network

Semi-supervised learning has received attention from researchers, as it allows one to exploit the structure of unlabeled data to achieve competitive classification results with much fewer labels than supervised approaches. The Local and Global Consistency (LGC) algorithm is one of the most well-known graph-based semi-supervised (GSSL) classifiers. Notably, its solution can be written as a linear combination of the known labels. The coefficients of this linear combination depend on a parameter𝛼, determining the decay of the reward over time when reaching labeled vertices in a random walk. In this work, we discuss how removing the self-influence of a labeled instance may be beneficial, and how it relates to leave-one-out error. Moreover, we propose to minimize this leave-one-out loss with automatic differentiation. Within this framework, we propose methods to estimate label reliability and diffusion rate. Optimizing the diffusion rate is more efficiently accomplished with a spectral representation. Results show that the label reliability approach competes with robust L1-norm methods and that removing diagonal entries reduces the risk of overfitting and leads to suitable criteria for parameter selection.

Optimizing Diffusion Rate and Label Reliability in a Graph-based Semi-supervised Classifier

Context: in the field of machine learning models are trained to learn from data, however often the context at which a model is deployed changes, degrading the performances of trained models and giving rise to a problem called Concept Drift (CD), which is a change in data distribution.Motivation: CD has attracted attention in machine learning literature, with works proposing modification to well-known algorithms’ structures, ensembles, online learning and drift detection, but most of the CD literature regards classification, while regression drift is still poorly explored. Objetive: The goal of this work is to perform a comparative study of CD detectors in the context of regression. Results: we found that (i) PH, KSWIN and EDDM showed higher detection averages; (ii) the base learner has a strong impact in CD detection and (iii) the rate at which CD happens also affects the detection process. Conclusion: our experiments were executed in a framework that can easily be extended to include new CD detectors and base learners, allowing future studies to use it.

A Comparative Study on Concept Drift Detectors for Regression

Automatic fingerprint identification systems (AFIS) are among the most used people identification solutions. As the size of fingerprint databases is continuously growing, studying fingerprint indexing mechanisms is desirable to facilitate the search process in a large-scale database. This work presents a method for fingerprint indexing, which uses both exact and approximation methods of nearest neighbors (ANNs), which are very efficient in terms of runtime, even if they sacrifice a little accuracy by presenting approximate solutions. In the presented approach, searches with ANN methods are made from deep embedding vectors extracted from image databases using a convolutional neural network (CNN). In this work, a CNN ResNet18 was used to extract the deep feature embeddings vectors, and the vectors vary in size between 64, 96, and 128. The ANNs methods tested for the query step were ANNOY, NGT, HNSW, and Nanoflann. The results were quite promising when using the FVC fingerprint databases (2000, 2002, and 2004), once we reached 100% hits in the searches with a penetration rate of 1%, with very low run times.

Deep Convolutional Features for Fingerprint Indexing

Reviews are valuable sources of information to support the decision making process. Therefore, the task of classifying reviews according to their helpfulness has paramount importance to facilitate the access of truly informative content. In this context, previous studies have unveiled several aspects and architectures that are beneficial for the task of review perceived helpfulness prediction. The present work aims to further investigate the influence of the text specificity aspect, defined as the level of details conveyed in a text, with the same purpose. First, we explore an unsupervised domain adaptation approach for assigning text specificity scores for sentences from product reviews and we propose an evaluation measure named Specificity Prediction Evaluation (SPE) in order to achieve more reliable specificity predictions. Then, we present domain-oriented guidelines on how to incorporate, into a CNN architecture, either hand-crafted features based on text specificity or the text specificity prediction task as an auxiliary task in a multitask learning setting. In the experiments, the perceived helpfulness classification models embodied with text specificity showed significant higher precision results in comparison to a popular SVM baseline.

Incorporating Text Specificity into a Convolutional Neural Network for the Classification of Review Perceived Helpfulness

The Federal Government Purchase Card (CPGF) has been used in Brazil since 2005, allowing agencies and entities of the federal public administration to make purchases of material and provision of services through this method. Although this payment system offers several advances, in the technological and administrative aspect, it is also susceptible to possible cases of card misuse and, consequently, waste of public funds, in the form of purchases that do not comply with the terms of the current legislation. In this work, we approach this problem by testing and evaluating unsupervised learning techniques on detecting anomalies in CPGF historical data. Four different methods are considered for this task: K-means, agglomerative clustering, a network-based approach, which is also introduced in this study, and a hybrid model. The experimental results obtained indicate that unsupervised methods, in particular the network-based approach, can indeed help in the task of monitoring government purchase card expenses, by flagging suspect transactions for further investigation without requiring the presence of a specialist in this process.

Anomaly Detection in Brazilian Federal Government Purchase Cards Through Unsupervised Learning Techniques

The popular participation in Law-making is an important resource in the evolution of Democracy and Direct Legislation. The amount of legislative documents produced within the past decade has risen dramatically, making it difficult for law practitioners to attend to legislation and still listen to the opinion of the citizens. This work focuses on the use of topic models for summarizing and visualizing Brazilian comments about legislation (bills). In this paper, we provide a qualitative evaluation from a legal expert and compare it with the topics predicted by our model. For such, we designed a specific sentence embedding technique able to induce models for Portuguese texts, and we used these models as topic model, obtaining very good results. We experimentally compared our proposal with other techniques for multilingual sentence embeddings, evaluating them in three topical corpora prepared by us, two of them annotated by a specialist and the other automatically annotated by hashtags.

Evaluating Topic Models in Portuguese Political Comments about Bills from Brazil's Chamber of Deputies

The dynamism of fake news evolution and dissemination plays a crucial role in influencing and confirming personal beliefs. To minimize the spread of disinformation approaches proposed in the literature, automatic fake news detection generally learns models through binary supervised algorithms considering textual and contextual information. However, labeling significant amounts of real news to build accurate classifiers is difficult and time-consuming due to their broad spectrum. Positive and unlabeled learning (PUL) can be a good alternative in this scenario. PUL algorithms learn models considering little labeled data of the interest class and use unlabeled data to increase classification performance. This paper proposes a heterogeneous network variant of the PU-LP algorithm, a PUL algorithm based on similarity networks. Our network incorporates different linguistic features to characterize fake news, such as representative terms, emotiveness, pausality, and average sentence size. Also, we considered two representations of the news to compute similarity: term frequency-inverse document frequency, and Doc2Vec, which creates fixed-sized document representations regardless of its length. We evaluated our approach in six datasets written in Portuguese or English, comparing its performance with a binary semi-supervised baseline algorithm, using two well-established label propagation algorithms: LPHN and GNetMine. The results indicate that PU-LP with heterogeneous networks can be competitive to binary semi-supervised learning. Also, linguistic features such as representative terms and pausality improved the classification performance, especially when there is a small amount of labeled news.

Ethics of AI: Do the face detection models act with prejudice?

Downloads

Next from BRACIS 2021

A Systematic Approach to Define Semantics for Prioritised Logic Programs

Similar lecture

An Introduction to Critical Race Theory for Physicists

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES