Questões: Instituto Federal de Educação, Ciência e Tecnologia do Sul de Minas Gerais (IF Sul - MG) — com gabarito

Assinale a alternativa INCORRETA a respeito dos movimentos literários brasileiros abaixo.

O texto abaixo é poético. Leia-o com atenção e resolva as questões.

“A lua,

tal qual a dona de um bordel,

pedia a cada estrela fria

um brilho de aluguel.

E nuvens,

lá no mata-borrão do céu

chupavam manchas torturadas

– que sufoco!”

(João Bosco e Aldir Blanc, “O bêbado e a equilibrista”)

I – No fragmento há predomínio da linguagem denotativa, característica máxima do texto literário.

II – A função referencial, presente no texto, transmite dados da realidade de forma subjetiva.

III – Nos textos literários, o autor explora determinadas construções com a intenção deliberada de reforçar a mensagem.

IV – No fragmento a figura de linguagem evidente é a prosopopeia.

Assinale a alternativa correta após análise completa.

Sobre a obra “Dois Irmãos” de Milton Hatoum, considere as seguintes afirmativas:

I – O título faz referência a dois irmãos que, supostamente, têm pais diferentes, nascendo daí uma rivalidade que destrói vidas e famílias.

II – O cenário principal da obra é a cidade de Manaus, valorizando o porto, as ruas do centro, as praças e os bairros mais antigos.

III – Rânia casou-se com o primeiro pretendente que apareceu para tentar esquecer o amor que sentia por Omar.

IV – Omar envolveu-se seriamente com duas mulheres: Dália e Pau-Mulato; porém nos dois casos a mãe dele interveio, provocando a separação.

V – Antes de morrer, Yaqub confessou que era o pai de Nael.

Assinale a alternativa que contém as afirmativas CORRETAS:

Assinale a alternativa INCORRETA sobre a obra “Dois Irmãos” de Miltom Hatoum.

Sobre o romance “Capitães da Areia”, de Jorge Amado, é INCORRETO afirmar que:

A evolução da comunicação é retratada com ironia no cartum. Com relação à figura e ao texto, é INCORRETO afirmar que:

Com relação às expressões abaixo, assinale a alternativa correta.

written word, movable type, mass publication

AI1 picks up racial and gender biases2 when learning language from text, researchers say. Without any supervision, a machine learning algorithm learns to associate female names more with family words than career words, and black names as being more unpleasant than white names.

For a study published today in Science, researchers tested the bias of a common AI model, and then matched the results against a well-known psychological test that measures bias in humans. The team replicated in the algorithm all the psychological biases they tested, according to a study from co-author Aylin Caliskan, a post-doc at Princeton University. Because machine learning algorithms are so common, influencing everything from translation to scanning names on resumes, this research shows that the biases are pervasive, too. 

An algorithm is a set of instructions that humans write to help computers learn. Think of it like a recipe, says Zachary Lipton, an AI researcher at UC San Diego who was not involved in the study. Because algorithms use existing materials — like books or text on the internet — it’s obvious that AI can pick up biases if the materials themselves are biased. (For example, Google Photos tagged black users as gorillas.) We’ve known for a while, for instance, that language algorithms learn to associate the word “man” with “professor” and the word “woman” with “assistant professor.” But this paper is interesting because it incorporates previous work done in psychology on human biases, Lipton says.

For today’s study, Caliskan’s team created a test that resembles the Implicit Association Test (IAT), which is commonly used in psychology to measure how biased people are (though there has been some controversy over its accuracy). In the IAT, subjects are presented with two images — say, a white man and a black man — and words like “pleasant” or “unpleasant.” The IAT calculates how quickly you match up “white man” and “pleasant” versus “black man” and “pleasant,” and vice versa. The idea is that the longer it takes you to match up two concepts, the more trouble you have associating them.

The test developed by the researchers also calculates bias, but instead of measuring “response time”, it measures the mathematical distance between two words. In other words, if there’s a bigger numerical distance between a black name and the concept of “pleasant” than a white name and “pleasant”, the model’s association between the two isn’t as strong. The further apart the words are, the less the algorithm associates them together.

Caliskan’s team then tested their method on one particular algorithm: Global Vectors for Word Representation (GLoVe) from Stanford University. GLoVe basically crawls the web to find data and learns associations between billions of words. The researchers found that, in GLoVe, female words are more associated with arts than with math or science, and black names are seen as more unpleasant than white names. That doesn’t mean there’s anything wrong with the AI system, per se, or how the AI is learning — there’s something wrong with the material.

Disponível em <http://www.theverge.com/>. Acesso em: 18/04/2017.

Assinale a alternativa que está de acordo com o texto.

AI1 picks up racial and gender biases2 when learning language from text, researchers say. Without any supervision, a machine learning algorithm learns to associate female names more with family words than career words, and black names as being more unpleasant than white names.

For a study published today in Science, researchers tested the bias of a common AI model, and then matched the results against a well-known psychological test that measures bias in humans. The team replicated in the algorithm all the psychological biases they tested, according to a study from co-author Aylin Caliskan, a post-doc at Princeton University. Because machine learning algorithms are so common, influencing everything from translation to scanning names on resumes, this research shows that the biases are pervasive, too. 

An algorithm is a set of instructions that humans write to help computers learn. Think of it like a recipe, says Zachary Lipton, an AI researcher at UC San Diego who was not involved in the study. Because algorithms use existing materials — like books or text on the internet — it’s obvious that AI can pick up biases if the materials themselves are biased. (For example, Google Photos tagged black users as gorillas.) We’ve known for a while, for instance, that language algorithms learn to associate the word “man” with “professor” and the word “woman” with “assistant professor.” But this paper is interesting because it incorporates previous work done in psychology on human biases, Lipton says.

For today’s study, Caliskan’s team created a test that resembles the Implicit Association Test (IAT), which is commonly used in psychology to measure how biased people are (though there has been some controversy over its accuracy). In the IAT, subjects are presented with two images — say, a white man and a black man — and words like “pleasant” or “unpleasant.” The IAT calculates how quickly you match up “white man” and “pleasant” versus “black man” and “pleasant,” and vice versa. The idea is that the longer it takes you to match up two concepts, the more trouble you have associating them.

The test developed by the researchers also calculates bias, but instead of measuring “response time”, it measures the mathematical distance between two words. In other words, if there’s a bigger numerical distance between a black name and the concept of “pleasant” than a white name and “pleasant”, the model’s association between the two isn’t as strong. The further apart the words are, the less the algorithm associates them together.

Caliskan’s team then tested their method on one particular algorithm: Global Vectors for Word Representation (GLoVe) from Stanford University. GLoVe basically crawls the web to find data and learns associations between billions of words. The researchers found that, in GLoVe, female words are more associated with arts than with math or science, and black names are seen as more unpleasant than white names. That doesn’t mean there’s anything wrong with the AI system, per se, or how the AI is learning — there’s something wrong with the material.

Disponível em <http://www.theverge.com/>. Acesso em: 18/04/2017.

Com relação ao teste desenvolvido pelos pesquisadores para calcular o preconceito, assinale a alternativa correta.

AI1 picks up racial and gender biases2 when learning language from text, researchers say. Without any supervision, a machine learning algorithm learns to associate female names more with family words than career words, and black names as being more unpleasant than white names.

For a study published today in Science, researchers tested the bias of a common AI model, and then matched the results against a well-known psychological test that measures bias in humans. The team replicated in the algorithm all the psychological biases they tested, according to a study from co-author Aylin Caliskan, a post-doc at Princeton University. Because machine learning algorithms are so common, influencing everything from translation to scanning names on resumes, this research shows that the biases are pervasive, too. 

An algorithm is a set of instructions that humans write to help computers learn. Think of it like a recipe, says Zachary Lipton, an AI researcher at UC San Diego who was not involved in the study. Because algorithms use existing materials — like books or text on the internet — it’s obvious that AI can pick up biases if the materials themselves are biased. (For example, Google Photos tagged black users as gorillas.) We’ve known for a while, for instance, that language algorithms learn to associate the word “man” with “professor” and the word “woman” with “assistant professor.” But this paper is interesting because it incorporates previous work done in psychology on human biases, Lipton says.

For today’s study, Caliskan’s team created a test that resembles the Implicit Association Test (IAT), which is commonly used in psychology to measure how biased people are (though there has been some controversy over its accuracy). In the IAT, subjects are presented with two images — say, a white man and a black man — and words like “pleasant” or “unpleasant.” The IAT calculates how quickly you match up “white man” and “pleasant” versus “black man” and “pleasant,” and vice versa. The idea is that the longer it takes you to match up two concepts, the more trouble you have associating them.

The test developed by the researchers also calculates bias, but instead of measuring “response time”, it measures the mathematical distance between two words. In other words, if there’s a bigger numerical distance between a black name and the concept of “pleasant” than a white name and “pleasant”, the model’s association between the two isn’t as strong. The further apart the words are, the less the algorithm associates them together.

Caliskan’s team then tested their method on one particular algorithm: Global Vectors for Word Representation (GLoVe) from Stanford University. GLoVe basically crawls the web to find data and learns associations between billions of words. The researchers found that, in GLoVe, female words are more associated with arts than with math or science, and black names are seen as more unpleasant than white names. That doesn’t mean there’s anything wrong with the AI system, per se, or how the AI is learning — there’s something wrong with the material.

Disponível em <http://www.theverge.com/>. Acesso em: 18/04/2017.

Why does Artificial Intelligence (AI) pick up biases?

A areia é um material muito utilizado na construção civil, sendo a unidade de medida utilizada para esse material o metro cúbico. Tião comprou um metro cúbico de areia, e esta foi entregue e depositada em sua calçada, obrigando-o a transportá-la para dentro de seu quintal. Para o transporte, Tião resolveu utilizar um caixote de madeira no formato de um cubo com 50 cm em cada lado. Assim, para transportar toda a areia da calçada para seu quintal, o número mínimo de viagens que Tião necessitará é de:

Essa divisão intrigava os herdeiros por não ser uma divisão exata. Supondo que fizessem a divisão por valores exatos, o número de camelos que sobrariam ainda para a partilha seria de:

Em uma turma de Sistemas de Informação formada por 30 rapazes e 20 moças tem-se a seguinte estatística: 20% dos rapazes são fumantes; 30% das moças são fumantes. Logo, a porcentagem dos que NÃO fumam nessa turma é de:

Uma famosa região rodeada por mistérios é o Triângulo das Bermudas, localizado entre a ilha Bermudas, Miami e San Juan (Porto Rico), que possuem como coordenadas aproximadas, respectivamente, (10,-4), (3,-20) e (-4,-6), dadas em graus (Latitude, Longitude). Considerando as coordenadas dadas para o Triângulo das Bermudas, a área dessa região em graus quadrados será de:

Em uma plantação de maracujás, as mudas são plantadas de forma alinhadas e a uma distância de 2m umas das outras. Como castigo a seu filho, um plantador de maracujás aplicou a seguinte regra:

Sabendo-se que a bica de água fica a 2m da primeira muda e que o copo deve ser levado até a bica ao fim do processo, esse fi lho caminhará em metros, entre idas e vindas, um total de:

Certo professor ganhou de presente uma caixa com 10 goiabas, sendo que 3 delas estavam verdes e as demais maduras. Retirando aleatoriamente duas goiabas da caixa (sem reposição), a probabilidade de serem retiradas duas goiabas maduras é:

Num “sapato de cromo”, o couro é tratado com um banho de “licor de cromo”, preparado por meio da reação representada pela equação:

Na2Cr2O7 + xSO2 + H2O → yCr(OH)SO4 + Na2 SO4

Ovos fertilizados de galinha foram bombardeados por meio da técnica de biobalística e passaram a expressar o gene LacZ oriundo de Escherichia coli, obtendo embriões transformados, capazes de produzir β-galactosidase, uma enzima utilizada na hidrólise da lactose.

Adaptado de In situ DNA transfer to chicken embryos by biolistics. RIBEIRO, Luciana A. et al. Disponível em <http://dx.doi.org/10.1590/S1415-47571999000400010>

Em janeiro de 2016, utilizando um dos solos da Terra mais parecidos com o de Marte (com até 20 vezes mais sais que a maior parte dos solos), pesquisadores peruanos e norte-americanos deram início ao projeto Batatas em Marte. O objetivo foi verificar se a batata pode ser cultivada em condições como as que existem no planeta vermelho. Embora ainda falte tratar de outros problemas, os primeiros resultados mostram que o tubérculo consegue crescer em solos tão áridos e salobros como os marcianos.

Adaptado de MIGUEL ÁNGEL 10/03/2017.Disponível em<http://brasil.elpais.com/brasil/2017/03/09/ciencia/1489084581_023882.html> .

Nessas condições ocorrerá, para fora das células da batata, o processo de: