计算机科学 ›› 2021, Vol. 48 ›› Issue (9): 1-8.doi: 10.11896/jsjkx.210600034
所属专题: 智能数据治理技术与系统
朝乐门, 尹显龙
CHAO Le-men, YIN Xian-long
摘要: 人工智能(Artificial Intelligence,AI)治理是解决AI挑战的主要手段。AI治理的主要目的是充分发挥人工智能带来的优势和有效降低人工智能导致的风险,并通过整合技术、法律、政策、标准、伦理、道德、安全、经济、社会等多个方面的影响因素,最终建设负责任的人工智能(Responsible Artificial Intelligence,RAI)。AI治理可以从智能个体治理、智能群体治理以及人机合作与共生系统的治理等3个方面,分技术层、伦理层、社会及法律层等3个层面进行。AI治理的主要关键技术有4种:可理解性人工智能、防御对抗性攻击技术、建模及仿真技术和实时审计技术。从谷歌、IBM和微软等公司的AI治理实践来看,产业界主要关注的是RAI研发,在AI系统的可解释性、隐私保护和公平性检查等方面已出现一些专用组件工具。目前,AI治理需要研究的科学问题有:软件定义的AI治理、AI治理关键技术、大规模机器学习中的AI治理评价、基于联邦学习的AI治理、AI治理的标准制定、增强人工智能与人在回路型AI训练等。
中图分类号:
[1]SHARMA G D,YADAV A,CHOPRA R.Artificial intelligence and effective governance:A review,critique and research agenda[J].Sustainable Futures,2020,2:100004. [2]WIRTZ B W,WEYERER J C,GEYER C.Artificial Intelligence and the Public Sector-Applications and Challenges[J].International Journal of Public Administration,2019,42(7):596-615. [3]AI governance:Ensuring your AI is transparent,compliant,and trustworthy [EB/OL].[2021-05-15].https://www.ibm.com/analytics/common/smartpapers/ai-governance-smartpaper/#ai-governance-delivers. [4]AI Governance:The path to responsibleadoption of artificial intelligence [R/OL].[2021-05-15].https://www.asianscientist.com/wp-content/uploads/2020/07/AI-Governance-Whitepaper-Basis-AI.pdf. [5]ULNICANE I,KNIGHT W,LEACH T,et al.Framing Gover-nance for a Contested Emerging Technology:Insights from AI Policy[J].Policy and Society,2020,40(2):1-20. [6]DURANTON S,MILLS S.Responsible AI:Leading by Example [EB/OL].(2021-02-03) [2021-05-15].https://medium.com/bcggamma/responsible-ai-leading-by-example-c25a8a0a98ea. [7]Responsible Machine Learning [EB/OL].[2021-05-15].https://www.h2o.ai/responsible-ai/. [8]WEARN O R,FREEMAN R,JACOBY D M.Responsible AIfor conservation[J].Nature Machine Intelligence,2019,1(2):72-73. [9]DAFOE A.AI governance:a research agenda[EB/OL].(2018-08-27)[2021-05-15].https://www.fhi.ox.ac.uk/wp-content/uploads/GovAI-Agenda.pdf. [10]KUZIEMSKI M,PALKA P.AI governance post-GDPR:lessons learned and the road ahead[J/OL].2019.http://diana-n.iue.it:8080/handle/1814/64146. [11]LI T,SAHU A K,TALWALKAR A,et al.Federated learning:Challenges,methods,and future directions[J].IEEE Signal Processing Magazine,2020,37(3):50-60. [12]GHALLAB M.Responsible AI:requirements and challenges[J].AI Perspectives,2019,1(1):1-7. [13]WIRTZ B W,WEYERER J C,STURM B J.The dark sides of artificial intelligence:An integrated AI governance framework for public administration[J].International Journal of Public Administration,2020,43(9):818-829. [14]GASSER U,ALMEIDA V A.A layered model for AI gover-nance[J].IEEE Internet Computing,2017,21(6):58-62. [15]ADLER S.From Data Governance to AI Governance:How to successfully make the shift? [EB/OL].[2021-05-15].https://www.aidataanalytics.network/data-science-ai/whitepapers/fromdata--governance-to-ai-governance-how-to-successfully-make-the-shift. [16]LEI Y,DUAN Y,SONG M.Technical Implementation Framework of AI Governance Policies for Cross-Modal Privacy Protection[C]//International Conference on Collaborative Computing:Networking,Applications and Worksharing.Springer,Cham,2020:431-443. [17]SCHIFF D,BIDDLE J,BORENSTEIN J,et al.What's Next for AI Ethics,Policy,and Governance? A Global Overview[C]//Proceedings of the AAAI/ACM Conference on AI,Ethics,and Society.2020. [18]THEODOROU A,DIGNUM V.Towards ethical and socio-legal governance in AI[J].Nature Machine Intelligence,2020,2(1):10-12. [19]WACHTER S,MITTELSTADT B,FLORIDI L.Transparent,explainable,and accountable AI for robotics[J].Science (Robo-tics),2017,2(6):1-5. [20]REDDY S,ALLAN S,COGHLAN S,et al.A governance model for the application of AI in health care[J].Journal of the American Medical Informatics Association,2020,27(3):491-497. [21]POMARES J,ABDALA M B.The future of AI governance [J/OL].[2021-05-15].https://www.global-solutions-initiative.org/wp-content/uploads/2020/04/GSJ5_Pomares_Abdala.pdf. [22]CIHON P,MAAS M M,KEMP L.Fragmentation and the Future:Investigating Architectures for International AI Gover-nance[J].Global Policy,2020,11(5):545-556. [23]CIHON P.Standards for AI governance:international standards to enable global coordination in AI research & development [R].Future of Humanity Institute University of Oxford.2019:1-41. [24]A practical guide to Responsible Artificial Intelligence (AI) [R/OL].[2021-05-15].https://www.pwc.com/gx/en/issues/data-and-analytics/artificial-intelligence/what-is-responsible-ai/responsible-ai-practical-guide.pdf. [25]GUNNING D,AHA D.DARPA's explainable artificial intelligence (XAI) program[J].AI Magazine,2019,40(2):44-58. [26]GUNNING D,STEFIK M,CHOI J,et al.XAI-Explainable artificial intelligence[J].Science Robotics,2019,4(37):1-2. [27]JIMÉNEZ-LUNA J,GRISONI F,SCHNEIDER G.Drug disco-very with explainable artificial intelligence[J].Nature Machine Intelligence,2020,2(10):573-584. [28]CHAKRABORTY A,ALAM M,DEY V,et al.Adversarial attacks anddefences:A survey[J].arXiv:181000069,2018. [29]YEUNG K,HOWES A,POGREBNA G.AI governance by human rights-centred design,deliberation and oversight:An end to ethics washing [M].The Oxford Handbook of AI Ethics,Oxford University Press,2019:1-27. [30]ZEIGLER B P,MUZY A,KOFMAN E.Theory of modeling and simulation:discrete event & iterative system computational foundations [M].Academic Press,2018. [31]ROTHROCK L,NARAYANAN S.Human-in-the-loop simulations [M].Springer,2011. [32]SALEIRO P,KUESTER B,HINKSON L,et al.Aequitas:A bias and fairness audit toolkit[J].arXiv:181105577,2018. [33]TORRIE V.AI Governance in Canadian Banking:Fairness,Credit Models,and Equality Rights[J].Credit Models,and Equality Rights,2020,36(1):5-38. [34]Implementation of the nationalAuroraAI programme [EB/OL].[2021-05-15].https://vm.fi/en/auroraai-en. [35]REMOLINA N,SEAH J.How to Address the AI Governance Discussion? What Can We LearnFrom Singapore's AI Strategy?[J].SMU Centre for AI & Data Governance Research Paper,2019(8):1-18. [36]SALEM F.A Smart City for public value:Digital transformation through agile governance-the case of ‘Smart Dubai'[J].World Government Summit Publications,Forthcoming,2020(5):1-70. [37]LEE D.TAIGER featured as an exemplary model for AI Ethics and Governance practices by IMDA and PDPC [EB/OL] (2020-10-19) [2021-05-15].https://taiger.com/articles/taiger-featured-as-an-exemplary-model-for-ai-ethics-and-governance-practices-by-imda-and-pdpc/. [38]AI PRINCIPLES OF TELEFÓNICA [EB/OL].[2021-05-15].https://www.telefonica.com/en/web/responsible -business/our-commitments/ai-principles. [39]Perspectives on Issues in AI Governance[R/OL].[2021-05-15].https://ai.google/static/documents/perspectives -on-issues-in-ai-governance.pdf. [40]AI Principles 2020 Progress update [R/OL].[2021-05-15].https://ai.google/static/documents/ai-principles-2020-progress-update.pdf. [41]PROST F,QIAN H,CHEN Q,et al.Toward a better trade-off between performance and fairness with kernel-based distribution matching[J].arXiv:191011779,2019. [42]WANG S,GUPTA M.Deontological ethics by monotonicityshape constraints[C]//International Conference on Artificial Intelligence and Statistics.PMLR,2020:2043-2054. [43]LAGE I,CHEN E,HE J,et al.Human evaluation of modelsbuilt for interpretability[C]//Proceedings of the AAAI Confe-rence on Human Computation and Crowdsourcing.2019. [44]KUCZMARSKI J.Reducing gender bias in Google Translate[EB/OL].(2018-12-6) [2021-05-15].https://blog.google/products/translate/reducing-gender-bias-google-translate/. [45]PFISFER T.Google Cloud,Harvard Global Health Institute release improved COVID-19 Public Forecasts,share lessons lear-ned [EB/OL].(November 17,2020) [2021-05-15].https://cloud.google.com/blog/products/ai-machine-learning/google-and-harvard-improve-covid-19-forecasts. [46]IBM,Artificial Intelligence [EB/OL].[2021-05-15].https://www.ibm.com/artificial-intelligence/ai-ethics-focus-areas. [47]TUCKER E,VAIDYANATHAN R.AI Governance:Drivecompliance,efficiency and outcomes from your AI lifecycle [EB/OL].(2020-05-26) [2021-05-15].https://www.ibm.com/blogs/journey-to-ai/2020/05/ai-governance-drive-compliance-ef-ficiency-and-outcomes-from-your-ai-lifecycle/?mhsrc=ibmsea-rch_a&mhq=AI-Governance. [48]VARSHNEY K R.Introducing AI Fairness 360 [EB/OL].(2018-09-19) [2021-05-15].https://www.ibm.com/blogs/research/2018/09/ai-fairness-360/. [49]MOJSILOVIC A.Introucing AI Explainability 360 [EB/OL].(2019-08-08) [2021-05-15].https://www.ibm.com/blogs/research/2019/08/ai-explainability-360/?mhsrc=ibmsearch_a& mhq=IBM%20 Explainability%20360. [50]HIND M.IBMFactSheets Further Advances Trust in AI[OL].(2020-07-09) [2021-05-15].https://www.ibm.com/blogs/research/2020/07/aifactsheets/?mhsrc=ibmsearch_a&mhq=AI%20Factsheet%20360. [51]SOKALSKI M.Artificial Intelligence in Control with Wat-sonOpenScale [EB/OL].[2021-05-15].https://www.kpmg.us/alliances/kpmg-ibm/ai-in-control-watson-openscale.html. [52]Research Collection:Research Supporting Responsible AI [EB/OL].(2020-04-13) [2021-05-15].https://www.microsoft.com/en-us/research/blog/research-collection-research-suppor-ting-responsible-ai/. [53]MADAIO M A,STARK L,WORTMAN V J,et al.Co-desig-ning checklists to understand organizational challenges and opportunities around fairness in ai [C]//Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems.2020. [54]GEBRU T,MORGENSTERN J,VECCHIONE B,et al.Data-sheets for datasets[J].arXiv:180309010,2018. [55]Responsible bots:10 guidelines for developers of conversational AI [OL].(2018-09) [2021-05-15].https://www.microsoft.com/en-us/research/publication/responsible-bots/. |
[1] | 丛颖男, 王兆毓, 朱金清. 关于法律人工智能数据和算法问题的若干思考 Insights into Dataset and Algorithm Related Problems in Artificial Intelligence for Law 计算机科学, 2022, 49(4): 74-79. https://doi.org/10.11896/jsjkx.210900191 |
[2] | 李野, 陈松灿. 基于物理信息的神经网络:最新进展与展望 Physics-informed Neural Networks:Recent Advances and Prospects 计算机科学, 2022, 49(4): 254-262. https://doi.org/10.11896/jsjkx.210500158 |
[3] | 景慧昀, 魏薇, 周川, 贺欣. 人工智能安全框架 Artificial Intelligence Security Framework 计算机科学, 2021, 48(7): 1-8. https://doi.org/10.11896/jsjkx.210300306 |
[4] | 谢宸琪, 张保稳, 易平. 人工智能模型水印研究综述 Survey on Artificial Intelligence Model Watermarking 计算机科学, 2021, 48(7): 9-16. https://doi.org/10.11896/jsjkx.201200204 |
[5] | 景慧昀, 周川, 贺欣. 针对人脸检测对抗攻击风险的安全测评方法 Security Evaluation Method for Risk of Adversarial Attack on Face Detection 计算机科学, 2021, 48(7): 17-24. https://doi.org/10.11896/jsjkx.210300305 |
[6] | 暴雨轩, 芦天亮, 杜彦辉, 石达. 基于i_ResNet34模型和数据增强的深度伪造视频检测方法 Deepfake Videos Detection Method Based on i_ResNet34 Model and Data Augmentation 计算机科学, 2021, 48(7): 77-85. https://doi.org/10.11896/jsjkx.210300258 |
[7] | 秦智慧, 李宁, 刘晓彤, 刘秀磊, 佟强, 刘旭红. 无模型强化学习研究综述 Overview of Research on Model-free Reinforcement Learning 计算机科学, 2021, 48(3): 180-187. https://doi.org/10.11896/jsjkx.200700217 |
[8] | 仝鑫, 王斌君, 王润正, 潘孝勤. 面向自然语言处理的深度学习对抗样本综述 Survey on Adversarial Sample of Deep Learning Towards Natural Language Processing 计算机科学, 2021, 48(1): 258-267. https://doi.org/10.11896/jsjkx.200500078 |
[9] | 周蔚, 罗旭东. 一种替代性纠纷在线仲裁系统 Alternative Online Arbitration System for Dispute 计算机科学, 2020, 47(6A): 583-590. https://doi.org/10.11896/JsJkx.190900140 |
[10] | 任仪. 基于区块链与人工智能的网络多服务器SIP信息加密系统设计 Design of Network Multi-server SIP Information Encryption System Based on Block Chain and Artificial Intelligence 计算机科学, 2020, 47(6A): 634-638. https://doi.org/10.11896/JsJkx.190600075 |
[11] | 赵澄, 叶耀威, 姚明海. 基于金融文本情感的股票波动预测 Stock Volatility Forecast Based on Financial Text Emotion 计算机科学, 2020, 47(5): 79-83. https://doi.org/10.11896/jsjkx.190400145 |
[12] | 王国胤, 瞿中, 赵显莲. 交叉融合的“人工智能+”学科建设探索与实践 Practical Exploration of Discipline Construction of Artificial Intelligence+ 计算机科学, 2020, 47(4): 1-5. https://doi.org/10.11896/jsjkx.200300144 |
[13] | 王晓明,赵歆波. 阅读眼动追踪语料库的构建与应用研究综述 Survey of Construction and Application of Reading Eye-tracking Corpus 计算机科学, 2020, 47(3): 174-181. https://doi.org/10.11896/jsjkx.190800040 |
[14] | 杨惟轶,白辰甲,蔡超,赵英男,刘鹏. 深度强化学习中稀疏奖励问题研究综述 Survey on Sparse Reward in Deep Reinforcement Learning 计算机科学, 2020, 47(3): 182-191. https://doi.org/10.11896/jsjkx.190200352 |
[15] | 曹锋,徐扬,钟建,宁欣然. 基于目标演绎距离的一阶逻辑子句集预处理方法 First-order Logic Clause Set Preprocessing Method Based on Goal Deduction Distance 计算机科学, 2020, 47(3): 217-221. https://doi.org/10.11896/jsjkx.190100004 |
|