Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31221
Scott Kuzdeba
The Internet of Things (IoT) has revolutionized how our devices are networked, connecting multiple aspects of our life from smart homes and wearables to smart cities and warehouses. IoT’s strength comes from the ever-expanding diverse heterogeneous sensors, applications, and concepts that are all centered around the core concept collecting and sharing data from sensors. Simultaneously, deep learning has changed how our systems operate, allowing them to learn from data and change the way we interface with the world. Federated learning moves these two paradigm shifts together, leveraging the data (securely) from the IoT to train deep learning architectures for performant edge applications. However, today’s federated learning has not yet benefited from the scale of diversity that the IoT and deep learning sensors and applications provide. This talk explores how we can better tap into the heterogeneity that surrounds the potential of federated learning and use it to build better models. This includes the heterogeneity from device hardware to training paradigms (supervised, unsupervised, reinforcement, self-supervised).
{"title":"Federated Learning of Things - Expanding the Heterogeneity in Federated Learning","authors":"Scott Kuzdeba","doi":"10.1609/aaaiss.v3i1.31221","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31221","url":null,"abstract":"The Internet of Things (IoT) has revolutionized how our devices are networked, connecting multiple\u0000aspects of our life from smart homes and wearables to smart cities and warehouses. IoT’s strength\u0000comes from the ever-expanding diverse heterogeneous sensors, applications, and concepts that are all\u0000centered around the core concept collecting and sharing data from sensors. Simultaneously, deep\u0000learning has changed how our systems operate, allowing them to learn from data and change the way\u0000we interface with the world. Federated learning moves these two paradigm shifts together, leveraging\u0000the data (securely) from the IoT to train deep learning architectures for performant edge applications. \u0000However, today’s federated learning has not yet benefited from the scale of diversity that the IoT and\u0000deep learning sensors and applications provide. This talk explores how we can better tap into the\u0000heterogeneity that surrounds the potential of federated learning and use it to build better models. This\u0000includes the heterogeneity from device hardware to training paradigms (supervised, unsupervised,\u0000reinforcement, self-supervised).","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"21 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141120646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31291
Bonan Zhao, Natalia Vélez, Thomas L. Griffiths
Human learning does not stop at solving a single problem. Instead, we seek new challenges, define new goals, and come up with new ideas. Unlike the classic explore-exploit trade-off between known and unknown options, making new tools or generating new ideas is not about collecting data from existing unknown options, but rather about create new options out of what is currently available. We introduce a discovery game designed to study how rational agents make decisions about pursuing innovations, where discovering new ideas is a process of combining existing ideas in an open-ended compositional space. We derive optimal policies of this decision problem formalized as a Markov decision process, and compare people's behaviors to the model predictions in an online behavioral experiment. We found evidence that people both innovate rationally, guided by potential returns in this discovery game, and under- and over-explore systematically in different settings.
{"title":"Comparing Human Behavior to an Optimal Policy for Innovation","authors":"Bonan Zhao, Natalia Vélez, Thomas L. Griffiths","doi":"10.1609/aaaiss.v3i1.31291","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31291","url":null,"abstract":"Human learning does not stop at solving a single problem. Instead, we seek new challenges, define new goals, and come up with new ideas. Unlike the classic explore-exploit trade-off between known and unknown options, making new tools or generating new ideas is not about collecting data from existing unknown options, but rather about create new options out of what is currently available. We introduce a discovery game designed to study how rational agents make decisions about pursuing innovations, where discovering new ideas is a process of combining existing ideas in an open-ended compositional space. We derive optimal policies of this decision problem formalized as a Markov decision process, and compare people's behaviors to the model predictions in an online behavioral experiment. We found evidence that people both innovate rationally, guided by potential returns in this discovery game, and under- and over-explore systematically in different settings.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"15 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141119915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31203
Kanak Raj, Kaushik Roy, Vamshi Bonagiri, Priyanshul Govil, K. Thirunarayan, Raxit Goswami, Manas Gaur
Personalizing conversational agents can enhance the quality of conversations and increase user engagement. However, they often lack external knowledge to appropriately tend to a user’s persona. This is crucial for practical applications like mental health support, nutrition planning, culturally sensitive conversations, or reducing toxic behavior in conversational agents. To enhance the relevance and comprehensiveness of personalized responses, we propose using a two-step approach that involves (1) selectively integrating user personas and (2) contextualizing the response by supplementing information from a background knowledge source. We develop K-PERM (Knowledge-guided PErsonalization with Reward Modulation), a dynamic conversational agent that combines these elements. K-PERM achieves state-of-the- art performance on the popular FoCus dataset, containing real-world personalized conversations concerning global landmarks.We show that using responses from K-PERM can improve performance in state-of-the-art LLMs (GPT 3.5) by 10.5%, highlighting the impact of K-PERM for personalizing chatbots.
{"title":"K-PERM: Personalized Response Generation Using Dynamic Knowledge Retrieval and Persona-Adaptive Queries","authors":"Kanak Raj, Kaushik Roy, Vamshi Bonagiri, Priyanshul Govil, K. Thirunarayan, Raxit Goswami, Manas Gaur","doi":"10.1609/aaaiss.v3i1.31203","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31203","url":null,"abstract":"Personalizing conversational agents can enhance the quality of conversations and increase user engagement. However, they often lack external knowledge to appropriately tend to a user’s persona. This is crucial for practical applications like mental health support, nutrition planning, culturally sensitive conversations, or reducing toxic behavior in conversational agents. To enhance the relevance and comprehensiveness of personalized responses, we propose using a two-step approach that involves (1) selectively integrating user personas and (2) contextualizing the response by supplementing information from a background knowledge source. We develop K-PERM (Knowledge-guided PErsonalization with Reward Modulation), a dynamic conversational agent that combines these elements. K-PERM achieves state-of-the- art performance on the popular FoCus dataset, containing real-world personalized conversations concerning global landmarks.We show that using responses from K-PERM can improve performance in state-of-the-art LLMs (GPT 3.5) by 10.5%, highlighting the impact of K-PERM for personalizing chatbots.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141120369","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31226
Nicholas Soures, Vedant Karia, D. Kudithipudi
Lifelong learning, which refers to an agent's ability to continuously learn and enhance its performance over its lifespan, is a significant challenge in artificial intelligence (AI), that biological systems tackle efficiently. This challenge is further exacerbated when AI is deployed in untethered environments with strict energy and latency constraints. We take inspiration from neural plasticity and investigate how to leverage and build energy-efficient lifelong learning machines. Specifically, we study how a combination of neural plasticity mechanisms, namely neuromodulation, synaptic consolidation, and metaplasticity, enhance the continual learning capabilities of AI models. We further co-design architectures that leverage compute-in-memory topologies and sparse spike-based communication with quantization for the edge. Aspects of this co-design can be transferred to federated lifelong learning scenarios.
{"title":"Advancing Neuro-Inspired Lifelong Learning for Edge with Co-Design","authors":"Nicholas Soures, Vedant Karia, D. Kudithipudi","doi":"10.1609/aaaiss.v3i1.31226","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31226","url":null,"abstract":"Lifelong learning, which refers to an agent's ability to continuously learn and enhance its performance over its lifespan, is a significant challenge in artificial intelligence (AI), that biological systems tackle efficiently. This challenge is further exacerbated when AI is deployed in untethered environments with strict energy and latency constraints. \u0000We take inspiration from neural plasticity and investigate how to leverage and build energy-efficient lifelong learning machines. Specifically, we study how a combination of neural plasticity mechanisms, namely neuromodulation, synaptic consolidation, and metaplasticity, enhance the continual learning capabilities of AI models. We further co-design architectures that leverage compute-in-memory topologies and sparse spike-based communication with quantization for the edge. Aspects of this co-design can be transferred to federated lifelong learning scenarios.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"15 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141120348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31292
Bonan Zhao, Christopher G Lucas, Neil R. Bramley
We propose bootstrap learning as a computational account for why human learning is modular and incremental, and identify key components of bootstrap learning that allow artificial systems to learn more like people. Originated from developmental psychology, bootstrap learning refers to people's ability to extend and repurpose existing knowledge to create new and more powerful ideas. We view bootstrap learning as a solution of how cognitively-bounded reasoners grasp complex environmental dynamics that are far beyond their initial capacity, by searching ‘locally’ and recursively to extend their existing knowledge. Drawing from techniques of Bayesian library learning and resource rational analysis, we propose a computational modeling framework that achieves human-like bootstrap learning performance in inductive conceptual inference. In addition, we demonstrate modeling and behavioral evidence that highlights the double-edged sword of bootstrap learning, such that people processing the same information in different batch orders could induce drastically different causal conclusions and generalizations, as a result of the different sub-concepts they construct in earlier stages of learning.
{"title":"Constructing Deep Concepts through Shallow Search","authors":"Bonan Zhao, Christopher G Lucas, Neil R. Bramley","doi":"10.1609/aaaiss.v3i1.31292","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31292","url":null,"abstract":"We propose bootstrap learning as a computational account for why human learning is modular and incremental, and identify key components of bootstrap learning that allow artificial systems to learn more like people. Originated from developmental psychology, bootstrap learning refers to people's ability to extend and repurpose existing knowledge to create new and more powerful ideas. We view bootstrap learning as a solution of how cognitively-bounded reasoners grasp complex environmental dynamics that are far beyond their initial capacity, by searching ‘locally’ and recursively to extend their existing knowledge. Drawing from techniques of Bayesian library learning and resource rational analysis, we propose a computational modeling framework that achieves human-like bootstrap learning performance in inductive conceptual inference. In addition, we demonstrate modeling and behavioral evidence that highlights the double-edged sword of bootstrap learning, such that people processing the same information in different batch orders could induce drastically different causal conclusions and generalizations, as a result of the different sub-concepts they construct in earlier stages of learning.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"40 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141118854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31182
Geoffrey W. Rutledge, Alexander Sivura
We describe "Dr. A.I.", a virtual physician assistant that uses generative AI to conduct a pre-visit patient interview and to create a draft clinical note for the physician. We document the effectiveness of Dr. A.I. by measuring the concordance of the actual diagnosis made by the doctor with the generated differ-ential diagnosis (DDx) list. This application demonstrates the practical healthcare capabilities of a large language model to improve efficiency of doctor visits while also addressing safety concerns for the use of generative AI in the workflow of patient care.
{"title":"A Generative AI-Based Virtual Physician Assistant","authors":"Geoffrey W. Rutledge, Alexander Sivura","doi":"10.1609/aaaiss.v3i1.31182","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31182","url":null,"abstract":"We describe \"Dr. A.I.\", a virtual physician assistant that uses generative AI to conduct a pre-visit patient interview and to create a draft clinical note for the physician. We document the effectiveness of Dr. A.I. by measuring the concordance of the actual diagnosis made by the doctor with the generated differ-ential diagnosis (DDx) list. This application demonstrates the practical healthcare capabilities of a large language model to improve efficiency of doctor visits while also addressing safety concerns for the use of generative AI in the workflow of patient care.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141120226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31250
K. Takadama
This paper explores an answer to the question of “what is a correct output by generative AI from the viewpoint of well-being?” and discusses an effectiveness of taking account of a biological rhythm for this issue. Concretely, this paper focuses on an estimation of the REM sleep stage as one of sleep stages, and compared its estimations based on random forest as one of the machine learning methods and the ultradian rhythm as one of the biological rhythms. From the human subject experiment, the following implications have been revealed: (1) the REM sleep stage is wrongly estimated in many areas by random forest; and (2) the integration of the REM sleep stage estimation based on the biological rhythm with that based on random forest improves the F-score of the estimated REM sleep stage.
本文探讨了 "从幸福的角度看，什么是生成式人工智能的正确输出？"这一问题的答案，并讨论了考虑生物节律对这一问题的有效性。具体而言，本文重点研究了作为睡眠阶段之一的快速眼动睡眠阶段的估算，并比较了基于随机森林（机器学习方法之一）和超昼夜节律（生物节律之一）的估算结果。通过人体实验，本文得出了以下结论：(1) 随机森林对快速动眼期睡眠阶段的估计在很多方面都是错误的；(2) 将基于生物节律的快速动眼期睡眠阶段估计与基于随机森林的快速动眼期睡眠阶段估计相结合，可以提高快速动眼期睡眠阶段估计的 F 分数。
{"title":"What Is a Correct Output by Generative AI From the Viewpoint of Well-Being? – Perspective From Sleep Stage Estimation –","authors":"K. Takadama","doi":"10.1609/aaaiss.v3i1.31250","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31250","url":null,"abstract":"This paper explores an answer to the question of “what is a correct output by generative AI from the viewpoint of well-being?” and discusses an effectiveness of taking account of a biological rhythm for this issue. Concretely, this paper focuses on an estimation of the REM sleep stage as one of sleep stages, and compared its estimations based on random forest as one of the machine learning methods and the ultradian rhythm as one of the biological rhythms. From the human subject experiment, the following implications have been revealed: (1) the REM sleep stage is wrongly estimated in many areas by random forest; and (2) the integration of the REM sleep stage estimation based on the biological rhythm with that based on random forest improves the F-score of the estimated REM sleep stage.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"71 14","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141121787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31254
Jin Yamanaka, Takashi Kido
After the significant performance of Large Language Models (LLMs) was revealed, their capabilities were rapidly expanded with techniques such as Retrieval Augmented Generation (RAG). Given their broad applicability and fast development, it's crucial to consider their impact on social systems. On the other hand, assessing these advanced LLMs poses challenges due to their extensive capabilities and the complex nature of social systems. In this study, we pay attention to the similarity between LLMs in social systems and humanoid robots in open environments. We enumerate the essential components required for controlling humanoids in problem solving which help us explore the core capabilities of LLMs and assess the effects of any deficiencies within these components. This approach is justified because the effectiveness of humanoid systems has been thoroughly proven and acknowledged. To identify needed components for humanoids in problem-solving tasks, we create an extensive component framework for planning and controlling humanoid robots in an open environment. Then assess the impacts and risks of LLMs for each component, referencing the latest benchmarks to evaluate their current strengths and weaknesses. Following the assessment guided by our framework, we identified certain capabilities that LLMs lack and concerns in social systems.
{"title":"Evaluating Large Language Models with RAG Capability: A Perspective from Robot Behavior Planning and Execution","authors":"Jin Yamanaka, Takashi Kido","doi":"10.1609/aaaiss.v3i1.31254","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31254","url":null,"abstract":"After the significant performance of Large Language Models (LLMs) was revealed, their capabilities were rapidly expanded with techniques such as Retrieval Augmented Generation (RAG). Given their broad applicability and fast development, it's crucial to consider their impact on social systems. On the other hand, assessing these advanced LLMs poses challenges due to their extensive capabilities and the complex nature of social systems.\u0000\u0000In this study, we pay attention to the similarity between LLMs in social systems and humanoid robots in open environments. We enumerate the essential components required for controlling humanoids in problem solving which help us explore the core capabilities of LLMs and assess the effects of any deficiencies within these components. This approach is justified because the effectiveness of humanoid systems has been thoroughly proven and acknowledged. To identify needed components for humanoids in problem-solving tasks, we create an extensive component framework for planning and controlling humanoid robots in an open environment. Then assess the impacts and risks of LLMs for each component, referencing the latest benchmarks to evaluate their current strengths and weaknesses. Following the assessment guided by our framework, we identified certain capabilities that LLMs lack and concerns in social systems.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"83 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141121105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31201
Ankur Padia, Francis Ferraro, Tim Finin
High-quality knowledge graphs (KGs) play a crucial role in many applications. However, KGs created by automated information extraction systems can suffer from erroneous extractions or be inconsistent with provenance/source text. It is important to identify and correct such problems. In this paper, we study leveraging the emergent reasoning capabilities of large language models (LLMs) to detect inconsistencies between extracted facts and their provenance. With a focus on ``open'' LLMs that can be run and trained locally, we find that few-shot approaches can yield an absolute performance gain of 2.5-3.4% over the state-of-the-art method with only 9% of training data. We examine the LLM architectures' effect and show that Decoder-Only models underperform Encoder-Decoder approaches. We also explore how model size impacts performance and counterintuitively find that larger models do not result in consistent performance gains. Our detailed analyses suggest that while LLMs can improve KG consistency, the different LLM models learn different aspects of KG consistency and are sensitive to the number of entities involved.
高质量的知识图谱（KG）在许多应用中发挥着至关重要的作用。然而，自动信息提取系统创建的知识图谱可能会出现提取错误或与出处/源文本不一致的情况。发现并纠正这些问题非常重要。在本文中，我们将研究如何利用大型语言模型（LLM）的新兴推理能力来检测提取事实与其出处之间的不一致性。我们将重点放在可在本地运行和训练的 "开放式 "LLM 上，结果发现，与最先进的方法相比，只需 9% 的训练数据，少数几种方法就能产生 2.5-3.4% 的绝对性能增益。我们研究了 LLM 架构的影响，结果表明仅解码器模型的性能低于编码器-解码器方法。我们还探讨了模型大小对性能的影响，并意外地发现较大的模型并不能带来一致的性能提升。我们的详细分析表明，虽然 LLM 可以提高 KG 一致性，但不同的 LLM 模型学习 KG 一致性的不同方面，并且对所涉及的实体数量很敏感。
{"title":"Enhancing Knowledge Graph Consistency through Open Large Language Models: A Case Study","authors":"Ankur Padia, Francis Ferraro, Tim Finin","doi":"10.1609/aaaiss.v3i1.31201","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31201","url":null,"abstract":"High-quality knowledge graphs (KGs) play a crucial role in many applications. However, KGs created by automated information extraction systems can suffer from erroneous extractions or be inconsistent with provenance/source text. It is important to identify and correct such problems. In this paper, we study leveraging the emergent reasoning capabilities of large language models (LLMs) to detect inconsistencies between extracted facts and their provenance. With a focus on ``open'' LLMs that can be run and trained locally, we find that few-shot approaches can yield an absolute performance gain of 2.5-3.4% over the state-of-the-art method with only 9% of training data. We examine the LLM architectures' effect and show that Decoder-Only models underperform Encoder-Decoder approaches. We also explore how model size impacts performance and counterintuitively find that larger models do not result in consistent performance gains. Our detailed analyses suggest that while LLMs can improve KG consistency, the different LLM models learn different aspects of KG consistency and are sensitive to the number of entities involved.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"23 19","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141119432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-20DOI: 10.1609/aaaiss.v3i1.31277
Maya Malaviya, Mark K. Ho
Humans are remarkably adaptive instructors who adjust advice based on their estimations about a learner’s prior knowledge and current goals. Many topics that people teach, like goal-directed behaviors, causal systems, categorization, and time-series patterns, have an underlying commonality: they map inputs to outputs through an unknown function. This project builds upon a Gaussian process (GP) regression model that describes learner behavior as they search the hypothesis space of possible underlying functions to find the one that best fits their current data. We extend this work by implementing a teacher model that reasons about a learner’s GP regression in order to provide specific information that will help them form an accurate estimation of the function.
人类是适应性极强的导师，他们会根据对学习者先前知识和当前目标的估计来调整建议。人类教授的许多主题，如目标导向行为、因果系统、分类和时间序列模式，都有一个潜在的共性：它们通过一个未知函数将输入映射到输出。本项目建立在高斯过程（GP）回归模型的基础上，该模型描述了学习者在搜索可能的基础函数的假设空间以找到最适合其当前数据的函数时的行为。我们通过实施一个教师模型来扩展这项工作，该模型可对学习者的 GP 回归进行推理，从而提供特定信息，帮助他们形成对函数的准确估计。
{"title":"Teaching Functions with Gaussian Process Regression","authors":"Maya Malaviya, Mark K. Ho","doi":"10.1609/aaaiss.v3i1.31277","DOIUrl":"https://doi.org/10.1609/aaaiss.v3i1.31277","url":null,"abstract":"Humans are remarkably adaptive instructors who adjust advice based on their estimations about a learner’s prior knowledge and current goals. Many topics that people teach, like goal-directed behaviors, causal systems, categorization, and time-series patterns, have an underlying commonality: they map inputs to outputs through an unknown function. This project builds upon a Gaussian process (GP) regression model that describes learner behavior as they search the hypothesis space of possible underlying functions to find the one that best fits their current data. We extend this work by implementing a teacher model that reasons about a learner’s GP regression in order to provide specific information that will help them form an accurate estimation of the function.","PeriodicalId":516827,"journal":{"name":"Proceedings of the AAAI Symposium Series","volume":"11 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141119812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：