\n
\n\n \n \n \n \n \n \n Digital Personhood? The Status of Autonomous Software Agents in Private Law.\n \n \n \n \n\n\n \n Teubner, G.\n\n\n \n\n\n\n
Ancilla Iuris, (106): 107–149. 2018.\n
ISSN: 1556-5068\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{teubner_digital_2018,\n\ttitle = {Digital {Personhood}? {The} {Status} of {Autonomous} {Software} {Agents} in {Private} {Law}},\n\tissn = {1661-8610},\n\tshorttitle = {Digital {Personhood}?},\n\turl = {https://www.jura.uni-frankfurt.de/71719886/Digital_PersonhoodENGancilla2018.pdf},\n\tdoi = {10.2139/ssrn.3177096},\n\tabstract = {Already today in the economy and in society, autonomous software agents, i.e. mathematically formalized information flows, are attributed social identity and ability to act under certain conditions. Due to social action attribution, they have become non-human members of society. They pose three new liability risks: (1) the autonomy risk, which has its origin in stand-alone “decisions” taken by the software agents, (2) the association risk, which is due to the close cooperation between people and software agents, and (3) the network risk that occurs when computer systems operate in close integration with other computer systems. These risks pose a challenge for private law: to define a new legal status for autonomous digital information systems—however not simply as complete legal personhood of software agents, human-computer associations or multi-agent systems respectively. Rather, in response to each of the three risks, a legal status should be granted to each of the algorithmic types that is carefully calibrated to their specific role. \nThree new forms of legal status for autonomous software agents are presented here: (1) actor with limited legal personhood, (2) member of a human-machine association, (3) element of a risk pool. For the risk of autonomy, an adequate response would be to recognize the software agents as legal actors with partial capability. Their autonomous choices should be legally binding, and these should, if found to be unlawful, trigger liability consequences. Software agents thereby are given a limited legal subjectivity, namely as representatives who conclude contracts for others. At the same time, they are to be recognized as legally capable persons in cases of contractual and non-contractual liability, so that the machine misbehavior itself—and not just the behavior of the underlying company—represents a breach of duty for which the company must stand. A possible answer to the association risk would be to give them legal status as a member of a human-machine association. The association itself would be recognized de lege ferenda as the legal subject of attribution for actions, rights and obligations. Finally, the answer to the network risk would be to define a risk pool, laid out autonomously according to liability law. The risk pool would define the legal status of the algorithms as part of a comprehensive digital information flow, with the liability of the pool resulting in the case of unlawful conduct of the pool.},\n\tlanguage = {en},\n\tnumber = {106},\n\turldate = {2025-02-11},\n\tjournal = {Ancilla Iuris},\n\tauthor = {Teubner, Gunther},\n\ttranslator = {Watson, Jacob},\n\tyear = {2018},\n\tnote = {ISSN: 1556-5068},\n\tpages = {107--149},\n}\n\n\n\n\n\n\n\n\n\n\n\n
\n
\n\n\n
\n Already today in the economy and in society, autonomous software agents, i.e. mathematically formalized information flows, are attributed social identity and ability to act under certain conditions. Due to social action attribution, they have become non-human members of society. They pose three new liability risks: (1) the autonomy risk, which has its origin in stand-alone “decisions” taken by the software agents, (2) the association risk, which is due to the close cooperation between people and software agents, and (3) the network risk that occurs when computer systems operate in close integration with other computer systems. These risks pose a challenge for private law: to define a new legal status for autonomous digital information systems—however not simply as complete legal personhood of software agents, human-computer associations or multi-agent systems respectively. Rather, in response to each of the three risks, a legal status should be granted to each of the algorithmic types that is carefully calibrated to their specific role. Three new forms of legal status for autonomous software agents are presented here: (1) actor with limited legal personhood, (2) member of a human-machine association, (3) element of a risk pool. For the risk of autonomy, an adequate response would be to recognize the software agents as legal actors with partial capability. Their autonomous choices should be legally binding, and these should, if found to be unlawful, trigger liability consequences. Software agents thereby are given a limited legal subjectivity, namely as representatives who conclude contracts for others. At the same time, they are to be recognized as legally capable persons in cases of contractual and non-contractual liability, so that the machine misbehavior itself—and not just the behavior of the underlying company—represents a breach of duty for which the company must stand. A possible answer to the association risk would be to give them legal status as a member of a human-machine association. The association itself would be recognized de lege ferenda as the legal subject of attribution for actions, rights and obligations. Finally, the answer to the network risk would be to define a risk pool, laid out autonomously according to liability law. The risk pool would define the legal status of the algorithms as part of a comprehensive digital information flow, with the liability of the pool resulting in the case of unlawful conduct of the pool.\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Adaptive Learning of Local Semantic and Global Structure Representations for Text Classification.\n \n \n \n \n\n\n \n Zhao, J.; Zhan, Z.; Yang, Q.; Zhang, Y.; Hu, C.; Li, Z.; Zhang, L.; and He, Z.\n\n\n \n\n\n\n In Bender, E. M.; Derczynski, L.; and Isabelle, P., editor(s),
Proceedings of the 27th International Conference on Computational Linguistics, pages 2033–2043, Santa Fe, New Mexico, USA, August 2018. Association for Computational Linguistics\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{zhao_adaptive_2018,\n\taddress = {Santa Fe, New Mexico, USA},\n\ttitle = {Adaptive {Learning} of {Local} {Semantic} and {Global} {Structure} {Representations} for {Text} {Classification}},\n\turl = {https://aclanthology.org/C18-1173/},\n\tabstract = {Representation learning is a key issue for most Natural Language Processing (NLP) tasks. Most existing representation models either learn little structure information or just rely on pre-defined structures, leading to degradation of performance and generalization capability. This paper focuses on learning both local semantic and global structure representations for text classification. In detail, we propose a novel Sandwich Neural Network (SNN) to learn semantic and structure representations automatically without relying on parsers. More importantly, semantic and structure information contribute unequally to the text representation at corpus and instance level. To solve the fusion problem, we propose two strategies: Adaptive Learning Sandwich Neural Network (AL-SNN) and Self-Attention Sandwich Neural Network (SA-SNN). The former learns the weights at corpus level, and the latter further combines attention mechanism to assign the weights at instance level. Experimental results demonstrate that our approach achieves competitive performance on several text classification tasks, including sentiment analysis, question type classification and subjectivity classification. Specifically, the accuracies are MR (82.1\\%), SST-5 (50.4\\%), TREC (96\\%) and SUBJ (93.9\\%).},\n\turldate = {2025-02-10},\n\tbooktitle = {Proceedings of the 27th {International} {Conference} on {Computational} {Linguistics}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Zhao, Jianyu and Zhan, Zhiqiang and Yang, Qichuan and Zhang, Yang and Hu, Changjian and Li, Zhensheng and Zhang, Liuxin and He, Zhiqiang},\n\teditor = {Bender, Emily M. and Derczynski, Leon and Isabelle, Pierre},\n\tmonth = aug,\n\tyear = {2018},\n\tpages = {2033--2043},\n}\n\n\n\n
\n
\n\n\n
\n Representation learning is a key issue for most Natural Language Processing (NLP) tasks. Most existing representation models either learn little structure information or just rely on pre-defined structures, leading to degradation of performance and generalization capability. This paper focuses on learning both local semantic and global structure representations for text classification. In detail, we propose a novel Sandwich Neural Network (SNN) to learn semantic and structure representations automatically without relying on parsers. More importantly, semantic and structure information contribute unequally to the text representation at corpus and instance level. To solve the fusion problem, we propose two strategies: Adaptive Learning Sandwich Neural Network (AL-SNN) and Self-Attention Sandwich Neural Network (SA-SNN). The former learns the weights at corpus level, and the latter further combines attention mechanism to assign the weights at instance level. Experimental results demonstrate that our approach achieves competitive performance on several text classification tasks, including sentiment analysis, question type classification and subjectivity classification. Specifically, the accuracies are MR (82.1%), SST-5 (50.4%), TREC (96%) and SUBJ (93.9%).\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Context-Aware Saliency Map Generation Using Semantic Segmentation.\n \n \n \n \n\n\n \n Ahmadi, M.; Hajabdollahi, M.; Karimi, N.; and Samavi, S.\n\n\n \n\n\n\n In
Electrical Engineering (ICEE), Iranian Conference on, pages 616–620, Mashhad, May 2018. IEEE\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{ahmadi_context-aware_2018,\n\taddress = {Mashhad},\n\ttitle = {Context-{Aware} {Saliency} {Map} {Generation} {Using} {Semantic} {Segmentation}},\n\tisbn = {978-1-5386-4914-5 978-1-5386-4916-9},\n\turl = {https://ieeexplore.ieee.org/document/8472577/},\n\tdoi = {10.1109/ICEE.2018.8472577},\n\turldate = {2025-02-10},\n\tbooktitle = {Electrical {Engineering} ({ICEE}), {Iranian} {Conference} on},\n\tpublisher = {IEEE},\n\tauthor = {Ahmadi, Mahdi and Hajabdollahi, Mohsen and Karimi, Nader and Samavi, Shadrokh},\n\tmonth = may,\n\tyear = {2018},\n\tpages = {616--620},\n}\n\n\n\n
\n
\n\n\n\n
\n
\n\n \n \n \n \n \n \n Digital Personhood? The Status of Autonomous Software Agents in Private Law.\n \n \n \n \n\n\n \n Teubner, G.\n\n\n \n\n\n\n May 2018.\n
\n\n
\n\n
\n\n
\n\n \n \n
Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@misc{teubner_digital_2018,\n\taddress = {Rochester, NY},\n\ttype = {{SSRN} {Scholarly} {Paper}},\n\ttitle = {Digital {Personhood}? {The} {Status} of {Autonomous} {Software} {Agents} in {Private} {Law}},\n\tshorttitle = {Digital {Personhood}?},\n\turl = {https://papers.ssrn.com/abstract=3177096},\n\tdoi = {10.2139/ssrn.3177096},\n\tabstract = {Already today in the economy and in society, autonomous software agents, i.e. mathematically formalized information flows, are attributed social identity and ability to act under certain conditions. Due to social action attribution, they have become non-human members of society. They pose three new liability risks: (1) the autonomy risk, which has its origin in stand-alone “decisions” taken by the software agents, (2) the association risk, which is due to the close cooperation between people and software agents, and (3) the network risk that occurs when computer systems operate in close integration with other computer systems. These risks pose a challenge for private law: to define a new legal status for autonomous digital information systems—however not simply as complete legal personhood of software agents, human-computer associations or multi-agent systems respectively. Rather, in response to each of the three risks, a legal status should be granted to each of the algorithmic types that is carefully calibrated to their specific role.},\n\tlanguage = {en},\n\turldate = {2024-11-18},\n\tpublisher = {Social Science Research Network},\n\tauthor = {Teubner, Gunther},\n\tmonth = may,\n\tyear = {2018},\n\tkeywords = {actant, digital personhood, hybrid, juridical person, legal person, liability, personification, software agents},\n}\n\n\n\n
\n
\n\n\n
\n Already today in the economy and in society, autonomous software agents, i.e. mathematically formalized information flows, are attributed social identity and ability to act under certain conditions. Due to social action attribution, they have become non-human members of society. They pose three new liability risks: (1) the autonomy risk, which has its origin in stand-alone “decisions” taken by the software agents, (2) the association risk, which is due to the close cooperation between people and software agents, and (3) the network risk that occurs when computer systems operate in close integration with other computer systems. These risks pose a challenge for private law: to define a new legal status for autonomous digital information systems—however not simply as complete legal personhood of software agents, human-computer associations or multi-agent systems respectively. Rather, in response to each of the three risks, a legal status should be granted to each of the algorithmic types that is carefully calibrated to their specific role.\n
\n\n\n