O Paradoxo da IA para Sustentabilidade e a Sustentabilidade da IA
Resumo
A popularização da inteligência artificial (IA) nos últimos anos tem gerado um impacto cada vez maior em diversos setores, fazendo com que seja necessária a análise das consequências de sua utilização frente a questões éticas e ambientais. Na área ambiental, pesquisas estão sendo realizadas no sentido de mensurar o impacto da utilização de algoritmos de IA em termos de consumo de energia e consequente emissão de dióxido de carbono equivalente (CO2e). Neste artigo, será abordado sobre o paradoxo envolvendo IA e sustentabilidade, com ênfase na importância de relatar o consumo de energia nas pesquisas envolvendo aprendizado de máquina(AM) e a viabilidade do uso de ferramentas online para realizar a medição da quantidade de CO2e emitida.
Referências
Anthony, L. F. W., Kanding, B., and Selvan, R. (2020). Carbontracker: Tracking and predicting the carbon footprint of training deep learning models. ICML Workshop on Challenges in Deploying and monitoring Machine Learning Systems. arXiv:2007.03051.
Assunção, V. D., Haniya, M. V., de Oliveira Fonseca, A. C., Jacob, A. C. P., Junior, J. T. A., de Magalhães, P. C., de Oliveira, A. K. B., and de Sousa, M. M. (2023). Análise dos desastres naturais em petrópolis ocorridos em fevereiro de 2022. ENCONTRO NACIONAL DE DESASTRES, 3.
Bender, E. M., Gebru, T., McMillan-Major, A., and Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, FAccT ’21, page 610–623, New York, NY, USA. Association for Computing Machinery.
Bouza, L., Bugeau, A., and Lannelongue, L. (2023). How to estimate carbon footprint when training deep learning models? a guide and review. Environmental Research Communications, 5.
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020). Language models are few-shot learners. arXiv preprint arXiv:2005.14165.
Budennyy, S. A., Lazarev, V. D., Zakharenko, N. N., Korovin, A. N., Plosskaya, O. A., Dimitrov, D. V., Akhripkin, V. S., Pavlov, I. V., Oseledets, I. V., Barsola, I. S., Egorov, I. V., Kosterina, A. A., and Zhukov, L. E. (2022). eco2ai: Carbon emissions tracking of machine learning models as the first step towards sustainable ai. Doklady Mathematics, 106(1):S118–S128.
Cowls, J., Tsamados, A., Taddeo, M., and Floridi, L. (2021). The ai gambit: leveraging artificial intelligence to combat climate change—opportunities, challenges, and recommendations. Ai & Society, pages 1 – 25.
Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL).
Ferro, M., Silva, G. D., de Paula, F. B., Vieira, V., and Schulze, B. (2023). Towards a sustainable artificial intelligence: A case study of energy efficiency in decision tree algorithms. Concurrency and Computation: Practice and Experience, 35(17):e6815.
Floridi, L. (2023). 209References. In The Ethics of Artificial Intelligence: Principles, Challenges, and Opportunities. Oxford University Press.
Henderson, P., Hu, J., Romoff, J., Brunskill, E., Jurafsky, D., and Pineau, J. (2020). Towards the systematic reporting of the energy and carbon footprints of machine learning.
Hennessy, J. L. and Patterson, D. A. (2011). Computer Architecture, Fifth Edition: A Quantitative Approach. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 5th edition.
Kaikkonen, L., Parviainen, T., Rahikainen, M., Uusitalo, L., and Lehikoinen, A. (2021). Bayesian networks in environmental risk assessment: A review. Integrated environmental assessment and management, 17(1):62–78.
Lacoste, A., Luccioni, A., Schmidt, V., and Dandres, T. (2019). Quantifying the carbon emissions of machine learning. arXiv preprint arXiv:1910.09700.
Lannelongue, L., Grealey, J., and Inouye, M. (2020). Green algorithms: Quantifying the carbon footprint of computation.
Li, P., Yang, J., Islam, M. A., and Ren, S. (2023). Making ai less ”thirsty”: Uncovering and addressing the secret water footprint of ai models.
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Baining, G. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. arXiv preprint arXiv:2103.14030.
Lottick, K., Susai, S., Friedler, S. A., and Wilson, J. P. (2019). Energy usage reports: Environmental awareness as part of algorithmic accountability.
Luccioni, A. S. and Hernandez-Garcia, A. (2023). Counting carbon: A survey of factors influencing the emissions of machine learning.
Luccioni, A. S., Viguier, S., and Ligozat, A.-L. (2022). Estimating the carbon footprint of bloom, a 176b parameter language model.
Machado, L., Gomes, M., Alves, J., and Freitas, F. (2023). Em 2022, mesmo depois da maior tragédia climática da história, petrópolis gastou apenas 15% do valor autorizado em habitação.
Mazzi, F. and Floridi, L., editors (2023). The Ethics of Artificial Intelligence for the Sustainable Development Goals. Springer Verlag.
Mehlin, V., Schacht, S., and Lanquillon, C. (2023). Towards energy-efficient deep learning: An overview of energy-efficient approaches along the deep learning lifecycle.
Mór, S., Alves, M., Lima, J., Maillard, N., and Navaux, P. (2015). Eficiência energética em computação de alto desempenho: Uma abordagem em arquitetura e programação para green computing.
Othman, A. (2023). Demystifying gpt and gpt-3: How they can support innovators to develop new digital accessibility solutions and assistive technologies? Nafath.
Pereira, R., Couto, M., Ribeiro, F., Rua, R., Cunha, J., Fernandes, J., and Saraiva, J. (2017). Energy efficiency across programming languages: how do energy, time, and memory relate? pages 256–267.
Porto, F., Ferro, M., Ogasawara, E., Moeda, T., Tenorio de Barros, C. D., Chaves Silva, A., Zorrilla, R., Silva Pereira, R., Nascimento Castro, R., Silva, J. V., Salles, R., Fonseca, A. J., Hermsdorff, J., Magalhães, M., Sá, V., Simões, A. A., Cardoso, C., and Bezerra, E. (2022). Machine learning approaches to extreme weather events forecast in urban areas: Challenges and initial results. Supercomputing Frontiers and Innovations, 9(1):49–73.
Ritchie, H. and Rosado, P. (2020). Electricity mix. Our World in Data. [link].
Schwartz, R., Dodge, J., Smith, N. A., and Etzioni, O. (2020). Green ai. Communications of the ACM, 63(12):54–63.
Souza, B. (2022). Petrópolis tem prejuízo de r$ 665 milhões e perde cerca de 2% do pib por consequência das chuvas.
Strubell, E., Ganesh, A., and McCallum, A. (2019). Energy and policy considerations for deep learning in nlp. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP).
Sætra, H. S. (2021). Ai in context and the sustainable development goals: Factoring in the unsustainability of the sociotechnical system. Sustainability, 13(4).
Thompson, N. C., Greenewald, K., Lee, K., and Manso, G. F. (2020). The computational limits of deep learning.
Tristan Trebaol, Mary-Anne Hartley, M. J. and Ghadikolaei, H. S. (2020). A tool to quantify and report the carbon footprint of machine learning computations and communication in academia and healthcare. Infoscience EPFL: record 278189.
UNESCO (2019). Preliminary study on the ethics of artificial intelligence.
UNESCO, D.-G. (2021). Preliminary report on the first draft of the recommendation on the ethics of artificial intelligence.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., and Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems. NIPS.
Vilaça, M. M., Pederneira, I. L., and Ferro, M. (2024). AI beyond a new academic hype: an interdisciplinary theoretical analytical experiment (computational, linguistic and ethical) of an AI tool. Unisinos Journal of Philosophy, 25:e25112.
Vinuesa, R., Azizpour, H., Leite, I., Balaam, M., Dignum, V., Domisch, S., Felländer, A., Langhans, S., Tegmark, M., and Nerini, F. F. (2020a). The role of AI in achieving the Sustainable Development Goals. Nature Communications, 11(233).
Vinuesa, R., Azizpour, H., Leite, I., Balaam, M., Dignum, V., Domisch, S., Felländer, A., Langhans, S. D., Tegmark, M., and Fuso Nerini, F. (2020b). The role of artificial intelligence in achieving the sustainable development goals. Nature Communications, 11(1):233.
Yang, T.-J., Chen, Y.-H., and Sze, V. (2017). Designing energy-efficient convolutional neural networks using energy-aware pruning. In arXiv preprint arXiv:1704.05155. Massachusetts Institute of Technology.
Yarally, T., Cruz, L., Feitosa, D., Sallou, J., and Van Deursen, A. (2023). Uncovering energy-efficient practices in deep learning training. In 2023 IEEE/ACM 2nd International Conference on AI Engineering – Software Engineering for AI (CAIN), pages 25–36. IEEE.
Yokoyama, A., Ferro, M., Paula, F., Vieira, V., and Schulze, B. (2023). Investigating hardware and software aspects in the energy consumption of machine learning: A green ai-centric analysis. Concurrency and Computation: Practice and Experience, 35.
Yusuf, M., Surana, P., Gupta, G., and Ramesh, K. (2022). Curb your carbon emissions: Benchmarking carbon emissions in machine translation. Preprint.
Zeng, A., Chen, M., Zhang, L., and Xu, Q. (2022). Are transformers effective for time series forecasting? In The Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI-23). AAAI Press.
Assunção, V. D., Haniya, M. V., de Oliveira Fonseca, A. C., Jacob, A. C. P., Junior, J. T. A., de Magalhães, P. C., de Oliveira, A. K. B., and de Sousa, M. M. (2023). Análise dos desastres naturais em petrópolis ocorridos em fevereiro de 2022. ENCONTRO NACIONAL DE DESASTRES, 3.
Bender, E. M., Gebru, T., McMillan-Major, A., and Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, FAccT ’21, page 610–623, New York, NY, USA. Association for Computing Machinery.
Bouza, L., Bugeau, A., and Lannelongue, L. (2023). How to estimate carbon footprint when training deep learning models? a guide and review. Environmental Research Communications, 5.
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020). Language models are few-shot learners. arXiv preprint arXiv:2005.14165.
Budennyy, S. A., Lazarev, V. D., Zakharenko, N. N., Korovin, A. N., Plosskaya, O. A., Dimitrov, D. V., Akhripkin, V. S., Pavlov, I. V., Oseledets, I. V., Barsola, I. S., Egorov, I. V., Kosterina, A. A., and Zhukov, L. E. (2022). eco2ai: Carbon emissions tracking of machine learning models as the first step towards sustainable ai. Doklady Mathematics, 106(1):S118–S128.
Cowls, J., Tsamados, A., Taddeo, M., and Floridi, L. (2021). The ai gambit: leveraging artificial intelligence to combat climate change—opportunities, challenges, and recommendations. Ai & Society, pages 1 – 25.
Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL).
Ferro, M., Silva, G. D., de Paula, F. B., Vieira, V., and Schulze, B. (2023). Towards a sustainable artificial intelligence: A case study of energy efficiency in decision tree algorithms. Concurrency and Computation: Practice and Experience, 35(17):e6815.
Floridi, L. (2023). 209References. In The Ethics of Artificial Intelligence: Principles, Challenges, and Opportunities. Oxford University Press.
Henderson, P., Hu, J., Romoff, J., Brunskill, E., Jurafsky, D., and Pineau, J. (2020). Towards the systematic reporting of the energy and carbon footprints of machine learning.
Hennessy, J. L. and Patterson, D. A. (2011). Computer Architecture, Fifth Edition: A Quantitative Approach. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, 5th edition.
Kaikkonen, L., Parviainen, T., Rahikainen, M., Uusitalo, L., and Lehikoinen, A. (2021). Bayesian networks in environmental risk assessment: A review. Integrated environmental assessment and management, 17(1):62–78.
Lacoste, A., Luccioni, A., Schmidt, V., and Dandres, T. (2019). Quantifying the carbon emissions of machine learning. arXiv preprint arXiv:1910.09700.
Lannelongue, L., Grealey, J., and Inouye, M. (2020). Green algorithms: Quantifying the carbon footprint of computation.
Li, P., Yang, J., Islam, M. A., and Ren, S. (2023). Making ai less ”thirsty”: Uncovering and addressing the secret water footprint of ai models.
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Baining, G. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. arXiv preprint arXiv:2103.14030.
Lottick, K., Susai, S., Friedler, S. A., and Wilson, J. P. (2019). Energy usage reports: Environmental awareness as part of algorithmic accountability.
Luccioni, A. S. and Hernandez-Garcia, A. (2023). Counting carbon: A survey of factors influencing the emissions of machine learning.
Luccioni, A. S., Viguier, S., and Ligozat, A.-L. (2022). Estimating the carbon footprint of bloom, a 176b parameter language model.
Machado, L., Gomes, M., Alves, J., and Freitas, F. (2023). Em 2022, mesmo depois da maior tragédia climática da história, petrópolis gastou apenas 15% do valor autorizado em habitação.
Mazzi, F. and Floridi, L., editors (2023). The Ethics of Artificial Intelligence for the Sustainable Development Goals. Springer Verlag.
Mehlin, V., Schacht, S., and Lanquillon, C. (2023). Towards energy-efficient deep learning: An overview of energy-efficient approaches along the deep learning lifecycle.
Mór, S., Alves, M., Lima, J., Maillard, N., and Navaux, P. (2015). Eficiência energética em computação de alto desempenho: Uma abordagem em arquitetura e programação para green computing.
Othman, A. (2023). Demystifying gpt and gpt-3: How they can support innovators to develop new digital accessibility solutions and assistive technologies? Nafath.
Pereira, R., Couto, M., Ribeiro, F., Rua, R., Cunha, J., Fernandes, J., and Saraiva, J. (2017). Energy efficiency across programming languages: how do energy, time, and memory relate? pages 256–267.
Porto, F., Ferro, M., Ogasawara, E., Moeda, T., Tenorio de Barros, C. D., Chaves Silva, A., Zorrilla, R., Silva Pereira, R., Nascimento Castro, R., Silva, J. V., Salles, R., Fonseca, A. J., Hermsdorff, J., Magalhães, M., Sá, V., Simões, A. A., Cardoso, C., and Bezerra, E. (2022). Machine learning approaches to extreme weather events forecast in urban areas: Challenges and initial results. Supercomputing Frontiers and Innovations, 9(1):49–73.
Ritchie, H. and Rosado, P. (2020). Electricity mix. Our World in Data. [link].
Schwartz, R., Dodge, J., Smith, N. A., and Etzioni, O. (2020). Green ai. Communications of the ACM, 63(12):54–63.
Souza, B. (2022). Petrópolis tem prejuízo de r$ 665 milhões e perde cerca de 2% do pib por consequência das chuvas.
Strubell, E., Ganesh, A., and McCallum, A. (2019). Energy and policy considerations for deep learning in nlp. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP).
Sætra, H. S. (2021). Ai in context and the sustainable development goals: Factoring in the unsustainability of the sociotechnical system. Sustainability, 13(4).
Thompson, N. C., Greenewald, K., Lee, K., and Manso, G. F. (2020). The computational limits of deep learning.
Tristan Trebaol, Mary-Anne Hartley, M. J. and Ghadikolaei, H. S. (2020). A tool to quantify and report the carbon footprint of machine learning computations and communication in academia and healthcare. Infoscience EPFL: record 278189.
UNESCO (2019). Preliminary study on the ethics of artificial intelligence.
UNESCO, D.-G. (2021). Preliminary report on the first draft of the recommendation on the ethics of artificial intelligence.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., and Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems. NIPS.
Vilaça, M. M., Pederneira, I. L., and Ferro, M. (2024). AI beyond a new academic hype: an interdisciplinary theoretical analytical experiment (computational, linguistic and ethical) of an AI tool. Unisinos Journal of Philosophy, 25:e25112.
Vinuesa, R., Azizpour, H., Leite, I., Balaam, M., Dignum, V., Domisch, S., Felländer, A., Langhans, S., Tegmark, M., and Nerini, F. F. (2020a). The role of AI in achieving the Sustainable Development Goals. Nature Communications, 11(233).
Vinuesa, R., Azizpour, H., Leite, I., Balaam, M., Dignum, V., Domisch, S., Felländer, A., Langhans, S. D., Tegmark, M., and Fuso Nerini, F. (2020b). The role of artificial intelligence in achieving the sustainable development goals. Nature Communications, 11(1):233.
Yang, T.-J., Chen, Y.-H., and Sze, V. (2017). Designing energy-efficient convolutional neural networks using energy-aware pruning. In arXiv preprint arXiv:1704.05155. Massachusetts Institute of Technology.
Yarally, T., Cruz, L., Feitosa, D., Sallou, J., and Van Deursen, A. (2023). Uncovering energy-efficient practices in deep learning training. In 2023 IEEE/ACM 2nd International Conference on AI Engineering – Software Engineering for AI (CAIN), pages 25–36. IEEE.
Yokoyama, A., Ferro, M., Paula, F., Vieira, V., and Schulze, B. (2023). Investigating hardware and software aspects in the energy consumption of machine learning: A green ai-centric analysis. Concurrency and Computation: Practice and Experience, 35.
Yusuf, M., Surana, P., Gupta, G., and Ramesh, K. (2022). Curb your carbon emissions: Benchmarking carbon emissions in machine translation. Preprint.
Zeng, A., Chen, M., Zhang, L., and Xu, Q. (2022). Are transformers effective for time series forecasting? In The Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI-23). AAAI Press.
Publicado
21/07/2024
Como Citar
BREDER, Gabriel B.; BRUM, Douglas F.; DIRK, Lucas; FERRO, Mariza.
O Paradoxo da IA para Sustentabilidade e a Sustentabilidade da IA. In: WORKSHOP SOBRE AS IMPLICAÇÕES DA COMPUTAÇÃO NA SOCIEDADE (WICS), 5. , 2024, Brasília/DF.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2024
.
p. 105-116.
ISSN 2763-8707.
DOI: https://doi.org/10.5753/wics.2024.2363.
