Pub Date : 2023-11-13DOI: 10.48550/arxiv.2311.07117
Chen, Kevin S., Pillow, Jonathan W., Leifer, Andrew M.
Animals adjust their behavioral response to sensory input adaptively depending on past experiences. The flexible brain computation is crucial for survival and is of great interest in neuroscience. The nematode C. elegans modulates its navigation behavior depending on the association of odor butanone with food (appetitive training) or starvation (aversive training), and will then climb up the butanone gradient or ignore it, respectively. However, the exact change in navigation strategy in response to learning is still unknown. Here we study the learned odor navigation in worms by combining precise experimental measurement and a novel descriptive model of navigation. Our model consists of two known navigation strategies in worms: biased random walk and weathervaning. We infer weights on these strategies by applying the model to worm navigation trajectories and the exact odor concentration it experiences. Compared to naive worms, appetitive trained worms up-regulate the biased random walk strategy, and aversive trained worms down-regulate the weathervaning strategy. The statistical model provides prediction with $>90 %$ accuracy of the past training condition given navigation data, which outperforms the classical chemotaxis metric. We find that the behavioral variability is altered by learning, such that worms are less variable after training compared to naive ones. The model further predicts the learning-dependent response and variability under optogenetic perturbation of the olfactory neuron AWC$^mathrm{ON}$. Lastly, we investigate neural circuits downstream from AWC$^mathrm{ON}$ that are differentially recruited for learned odor-guided navigation. Together, we provide a new paradigm to quantify flexible navigation algorithms and pinpoint the underlying neural substrates.
{"title":"Olfactory learning alters navigation strategies and behavioral\u0000 variability in C. elegans","authors":"Chen, Kevin S., Pillow, Jonathan W., Leifer, Andrew M.","doi":"10.48550/arxiv.2311.07117","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07117","url":null,"abstract":"Animals adjust their behavioral response to sensory input adaptively depending on past experiences. The flexible brain computation is crucial for survival and is of great interest in neuroscience. The nematode C. elegans modulates its navigation behavior depending on the association of odor butanone with food (appetitive training) or starvation (aversive training), and will then climb up the butanone gradient or ignore it, respectively. However, the exact change in navigation strategy in response to learning is still unknown. Here we study the learned odor navigation in worms by combining precise experimental measurement and a novel descriptive model of navigation. Our model consists of two known navigation strategies in worms: biased random walk and weathervaning. We infer weights on these strategies by applying the model to worm navigation trajectories and the exact odor concentration it experiences. Compared to naive worms, appetitive trained worms up-regulate the biased random walk strategy, and aversive trained worms down-regulate the weathervaning strategy. The statistical model provides prediction with $>90 %$ accuracy of the past training condition given navigation data, which outperforms the classical chemotaxis metric. We find that the behavioral variability is altered by learning, such that worms are less variable after training compared to naive ones. The model further predicts the learning-dependent response and variability under optogenetic perturbation of the olfactory neuron AWC$^mathrm{ON}$. Lastly, we investigate neural circuits downstream from AWC$^mathrm{ON}$ that are differentially recruited for learned odor-guided navigation. Together, we provide a new paradigm to quantify flexible navigation algorithms and pinpoint the underlying neural substrates.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352471","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}
Text-rich VQA, namely Visual Question Answering based on text recognition in the images, is a cross-modal task that requires both image comprehension and text recognition. In this work, we focus on investigating the advantages and bottlenecks of LLM-based approaches in addressing this problem. To address the above concern, we separate the vision and language modules, where we leverage external OCR models to recognize texts in the image and Large Language Models (LLMs) to answer the question given texts. The whole framework is training-free benefiting from the in-context ability of LLMs. This pipeline achieved superior performance compared to the majority of existing Multimodal Large Language Models (MLLM) on four text-rich VQA datasets. Besides, based on the ablation study, we find that LLM brings stronger comprehension ability and may introduce helpful knowledge for the VQA problem. The bottleneck for LLM to address text-rich VQA problems may primarily lie in visual part. We also combine the OCR module with MLLMs and pleasantly find that the combination of OCR module with MLLM also works. It's worth noting that not all MLLMs can comprehend the OCR information, which provides insights into how to train an MLLM that preserves the abilities of LLM.
富文本的VQA，即基于图像文本识别的视觉问答，是一项既需要图像理解又需要文本识别的跨模态任务。在这项工作中，我们重点研究了基于法学硕士的方法在解决这个问题方面的优势和瓶颈。为了解决上述问题，我们分离了视觉和语言模块，我们利用外部OCR模型来识别图像中的文本，并利用大型语言模型(llm)来回答给定文本的问题。整个框架是免培训的，受益于法学硕士的上下文能力。在四个文本丰富的VQA数据集上，与大多数现有的多模态大型语言模型(Multimodal Large Language Models, MLLM)相比，该管道取得了卓越的性能。此外，通过消融研究，我们发现LLM带来了更强的理解能力，可以为VQA问题引入有用的知识。LLM解决文本丰富的VQA问题的瓶颈可能主要在于可视化部分。我们还将OCR模块与MLLM相结合，并愉快地发现OCR模块与MLLM相结合也是有效的。值得注意的是，并非所有的mlm都能理解OCR信息，这为如何训练一个保留LLM能力的mlm提供了见解。
{"title":"What Large Language Models Bring to Text-rich VQA?","authors":"Liu, Xuejing, Tang, Wei, Ni, Xinzhe, Lu, Jinghui, Zhao, Rui, Li, Zechao, Tan, Fei","doi":"10.48550/arxiv.2311.07306","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07306","url":null,"abstract":"Text-rich VQA, namely Visual Question Answering based on text recognition in the images, is a cross-modal task that requires both image comprehension and text recognition. In this work, we focus on investigating the advantages and bottlenecks of LLM-based approaches in addressing this problem. To address the above concern, we separate the vision and language modules, where we leverage external OCR models to recognize texts in the image and Large Language Models (LLMs) to answer the question given texts. The whole framework is training-free benefiting from the in-context ability of LLMs. This pipeline achieved superior performance compared to the majority of existing Multimodal Large Language Models (MLLM) on four text-rich VQA datasets. Besides, based on the ablation study, we find that LLM brings stronger comprehension ability and may introduce helpful knowledge for the VQA problem. The bottleneck for LLM to address text-rich VQA problems may primarily lie in visual part. We also combine the OCR module with MLLMs and pleasantly find that the combination of OCR module with MLLM also works. It's worth noting that not all MLLMs can comprehend the OCR information, which provides insights into how to train an MLLM that preserves the abilities of LLM.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352485","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 : 2023-11-13DOI: 10.48550/arxiv.2311.07469
Friedl, Ken E., Khan, Abbas Goher, Sahoo, Soumya Ranjan, Rony, Md Rashad Al Hasan, Germies, Jana, Süß, Christian
The assessment of advanced generative large language models (LLMs) poses a significant challenge, given their heightened complexity in recent developments. Furthermore, evaluating the performance of LLM-based applications in various industries, as indicated by Key Performance Indicators (KPIs), is a complex undertaking. This task necessitates a profound understanding of industry use cases and the anticipated system behavior. Within the context of the automotive industry, existing evaluation metrics prove inadequate for assessing in-car conversational question answering (ConvQA) systems. The unique demands of these systems, where answers may relate to driver or car safety and are confined within the car domain, highlight the limitations of current metrics. To address these challenges, this paper introduces a set of KPIs tailored for evaluating the performance of in-car ConvQA systems, along with datasets specifically designed for these KPIs. A preliminary and comprehensive empirical evaluation substantiates the efficacy of our proposed approach. Furthermore, we investigate the impact of employing varied personas in prompts and found that it enhances the model's capacity to simulate diverse viewpoints in assessments, mirroring how individuals with different backgrounds perceive a topic.
{"title":"InCA: Rethinking In-Car Conversational System Assessment Leveraging\u0000 Large Language Models","authors":"Friedl, Ken E., Khan, Abbas Goher, Sahoo, Soumya Ranjan, Rony, Md Rashad Al Hasan, Germies, Jana, Süß, Christian","doi":"10.48550/arxiv.2311.07469","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07469","url":null,"abstract":"The assessment of advanced generative large language models (LLMs) poses a significant challenge, given their heightened complexity in recent developments. Furthermore, evaluating the performance of LLM-based applications in various industries, as indicated by Key Performance Indicators (KPIs), is a complex undertaking. This task necessitates a profound understanding of industry use cases and the anticipated system behavior. Within the context of the automotive industry, existing evaluation metrics prove inadequate for assessing in-car conversational question answering (ConvQA) systems. The unique demands of these systems, where answers may relate to driver or car safety and are confined within the car domain, highlight the limitations of current metrics. To address these challenges, this paper introduces a set of KPIs tailored for evaluating the performance of in-car ConvQA systems, along with datasets specifically designed for these KPIs. A preliminary and comprehensive empirical evaluation substantiates the efficacy of our proposed approach. Furthermore, we investigate the impact of employing varied personas in prompts and found that it enhances the model's capacity to simulate diverse viewpoints in assessments, mirroring how individuals with different backgrounds perceive a topic.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352512","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}
The neutron rich neck zone created in heavy ion reaction is experimentally probed by the production of the $A=3$ isobars. The energy spectra and angular distributions of triton and $^3$He are measured with the CSHINE detector in $^{86}$Kr +$^{208}$Pb reactions at 25 MeV/u. While the energy spectrum of $^{3}$He is harder than that of triton, known as "$^{3}$He-puzzle", the yield ratio $R({rm t/^3He})$ presents a robust rising trend with the polar angle in laboratory. Using the fission fragments to reconstruct the fission plane, the enhancement of out-plane $R({rm t/^3He})$ is confirmed in comparison to the in-plane ratios. Transport model simulations reproduce qualitatively the experimental trends, but the quantitative agreement is not achieved. The results demonstrate that a neutron rich neck zone is formed in the reactions. Further studies are called for to understand the clustering and the isospin dynamics related to neck formation.
{"title":"Revisit to the yield ratio of triton and $^3$He as an indicator of\u0000 neutron-rich neck emission","authors":"Wang, Yijie, Wan, Mengting, Diao, Xinyue, Xiao, Sheng, Qin, Yuhao, Qin, Zhi, Guo, Dong, Si, Dawei, Zhang, Boyuan, Tian, Baiting, Guan, Fenhai, Wu, Qianghua, Wei, Xianglun, Yang, Herun, Ma, Peng, Hu, Rongjiang, Duan, Limin, Duan, Fangfang, Ma, Junbing, Xu, Shiwei, Hu, Qiang, Bai, Zhen, Yang, Yanyun, Wang, Jiansong, Liu, Wenbo, Su, Wanqing, Wei, Xiaobao, Ma, Chunwang, Li, Xinxiang, Wang, Hongwei, Zhang, Yingxun, Warda, Michał, Dobrowolski, Arthur, Nerlo-Pomorska, Bożena, Pomorski, Krzysztof, Ou, Li, Xiao, Zhigang","doi":"10.48550/arxiv.2311.07095","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07095","url":null,"abstract":"The neutron rich neck zone created in heavy ion reaction is experimentally probed by the production of the $A=3$ isobars. The energy spectra and angular distributions of triton and $^3$He are measured with the CSHINE detector in $^{86}$Kr +$^{208}$Pb reactions at 25 MeV/u. While the energy spectrum of $^{3}$He is harder than that of triton, known as \"$^{3}$He-puzzle\", the yield ratio $R({rm t/^3He})$ presents a robust rising trend with the polar angle in laboratory. Using the fission fragments to reconstruct the fission plane, the enhancement of out-plane $R({rm t/^3He})$ is confirmed in comparison to the in-plane ratios. Transport model simulations reproduce qualitatively the experimental trends, but the quantitative agreement is not achieved. The results demonstrate that a neutron rich neck zone is formed in the reactions. Further studies are called for to understand the clustering and the isospin dynamics related to neck formation.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352631","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}
(Source) code search is widely concerned by software engineering researchers because it can improve the productivity and quality of software development. Given a functionality requirement usually described in a natural language sentence, a code search system can retrieve code snippets that satisfy the requirement from a large-scale code corpus, e.g., GitHub. To realize effective and efficient code search, many techniques have been proposed successively. These techniques improve code search performance mainly by optimizing three core components, including query understanding component, code understanding component, and query-code matching component. In this paper, we provide a 3-dimensional perspective survey for code search. Specifically, we categorize existing code search studies into query-end optimization techniques, code-end optimization techniques, and match-end optimization techniques according to the specific components they optimize. Considering that each end can be optimized independently and contributes to the code search performance, we treat each end as a dimension. Therefore, this survey is 3-dimensional in nature, and it provides a comprehensive summary of each dimension in detail. To understand the research trends of the three dimensions in existing code search studies, we systematically review 68 relevant literatures. Different from existing code search surveys that only focus on the query end or code end or introduce various aspects shallowly (including codebase, evaluation metrics, modeling technique, etc.), our survey provides a more nuanced analysis and review of the evolution and development of the underlying techniques used in the three ends. Based on a systematic review and summary of existing work, we outline several open challenges and opportunities at the three ends that remain to be addressed in future work.
{"title":"A Survey of Source Code Search: A 3-Dimensional Perspective","authors":"Sun, Weisong, Fang, Chunrong, Ge, Yifei, Hu, Yuling, Chen, Yuchen, Zhang, Quanjun, Ge, Xiuting, Liu, Yang, Chen, Zhenyu","doi":"10.48550/arxiv.2311.07107","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07107","url":null,"abstract":"(Source) code search is widely concerned by software engineering researchers because it can improve the productivity and quality of software development. Given a functionality requirement usually described in a natural language sentence, a code search system can retrieve code snippets that satisfy the requirement from a large-scale code corpus, e.g., GitHub. To realize effective and efficient code search, many techniques have been proposed successively. These techniques improve code search performance mainly by optimizing three core components, including query understanding component, code understanding component, and query-code matching component. In this paper, we provide a 3-dimensional perspective survey for code search. Specifically, we categorize existing code search studies into query-end optimization techniques, code-end optimization techniques, and match-end optimization techniques according to the specific components they optimize. Considering that each end can be optimized independently and contributes to the code search performance, we treat each end as a dimension. Therefore, this survey is 3-dimensional in nature, and it provides a comprehensive summary of each dimension in detail. To understand the research trends of the three dimensions in existing code search studies, we systematically review 68 relevant literatures. Different from existing code search surveys that only focus on the query end or code end or introduce various aspects shallowly (including codebase, evaluation metrics, modeling technique, etc.), our survey provides a more nuanced analysis and review of the evolution and development of the underlying techniques used in the three ends. Based on a systematic review and summary of existing work, we outline several open challenges and opportunities at the three ends that remain to be addressed in future work.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352647","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 : 2023-11-13DOI: 10.48550/arxiv.2311.07281
Schießl, Jonas, Baumann, Michael H., Faulwasser, Timm, Grüne, Lars
In this paper, we introduce and study different dissipativity notions and different turnpike properties for discrete-time stochastic nonlinear optimal control problems. The proposed stochastic dissipativity notions extend the classic notion of Jan C. Willems to $L^r$ random variables and to probability measures. Our stochastic turnpike properties range from a formulation for random variables via turnpike phenomena in probability and in probability measures to the turnpike property for the moments. Moreover, we investigate how different metrics (such as Wasserstein or L'evy-Prokhorov) can be leveraged in the analysis. Our results are built upon stationarity concepts in distribution and in random variables and on the formulation of the stochastic optimal control problem as a finite-horizon Markov decision process. We investigate how the proposed dissipativity notions connect to the various stochastic turnpike properties and we work out the link between these two different forms of dissipativity.
本文引入并研究了离散时间随机非线性最优控制问题的不同耗散率概念和不同收费公路性质。提出的随机耗散率概念将Jan C. Willems的经典概念扩展到L^r$随机变量和概率测度。我们的随机收费公路性质的范围从随机变量的公式通过收费公路现象的概率和概率措施的收费公路性质的矩。此外，我们还研究了如何在分析中利用不同的指标(如Wasserstein或L'evy-Prokhorov)。我们的结果建立在分布和随机变量的平稳性概念以及作为有限视界马尔可夫决策过程的随机最优控制问题的公式的基础上。我们研究了所提出的耗散率概念如何与各种随机收费公路性质联系起来，并找出了这两种不同形式的耗散率之间的联系。
{"title":"On the relationship between stochastic turnpike and dissipativity\u0000 notions","authors":"Schießl, Jonas, Baumann, Michael H., Faulwasser, Timm, Grüne, Lars","doi":"10.48550/arxiv.2311.07281","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07281","url":null,"abstract":"In this paper, we introduce and study different dissipativity notions and different turnpike properties for discrete-time stochastic nonlinear optimal control problems. The proposed stochastic dissipativity notions extend the classic notion of Jan C. Willems to $L^r$ random variables and to probability measures. Our stochastic turnpike properties range from a formulation for random variables via turnpike phenomena in probability and in probability measures to the turnpike property for the moments. Moreover, we investigate how different metrics (such as Wasserstein or L'evy-Prokhorov) can be leveraged in the analysis. Our results are built upon stationarity concepts in distribution and in random variables and on the formulation of the stochastic optimal control problem as a finite-horizon Markov decision process. We investigate how the proposed dissipativity notions connect to the various stochastic turnpike properties and we work out the link between these two different forms of dissipativity.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352649","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 : 2023-11-13DOI: 10.48550/arxiv.2311.07255
Parviainen, Hannu, Luque, Rafael, Palle, Enric
We present Spright, a Python package that implements a fast and lightweight mass-density-radius relation for small planets. The relation represents the joint planetary radius and bulk density probability distribution as a mean posterior predictive distribution of an analytical three-component mixture model. The analytical model, in turn, represents the probability for the planetary bulk density as three generalised Student's t-distributions with radius-dependent weights and means based on theoretical composition models. The approach is based on Bayesian inference and aims to overcome the rigidity of simple parametric mass-radius relations and the danger of overfitting of non-parametric mass-radius relations. The package includes a set of pre-trained and ready-to-use relations based on two M dwarf catalogues, one FGK star catalogue, and two theoretical composition models for water-rich planets. The inference of new models is easy and fast, and the package includes a command line tool that allows for coding-free use of the relation, including the creation of publication-quality plots. Additionally, we study whether the current mass and radius observations of small exoplanets support the presence of a population of water-rich planets positioned between rocky planets and sub-Neptunes. The study is based on Bayesian model comparison and shows somewhat strong support against the existence of a water-world population around M dwarfs. However, the results of the study depend on the chosen theoretical water-world density model. A more conclusive result requires a larger sample of precisely characterised planets and community consensus on a realistic water world interior structure and atmospheric composition model.
{"title":"Spright: a probabilistic mass-density-radius relation for small planets","authors":"Parviainen, Hannu, Luque, Rafael, Palle, Enric","doi":"10.48550/arxiv.2311.07255","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07255","url":null,"abstract":"We present Spright, a Python package that implements a fast and lightweight mass-density-radius relation for small planets. The relation represents the joint planetary radius and bulk density probability distribution as a mean posterior predictive distribution of an analytical three-component mixture model. The analytical model, in turn, represents the probability for the planetary bulk density as three generalised Student's t-distributions with radius-dependent weights and means based on theoretical composition models. The approach is based on Bayesian inference and aims to overcome the rigidity of simple parametric mass-radius relations and the danger of overfitting of non-parametric mass-radius relations. The package includes a set of pre-trained and ready-to-use relations based on two M dwarf catalogues, one FGK star catalogue, and two theoretical composition models for water-rich planets. The inference of new models is easy and fast, and the package includes a command line tool that allows for coding-free use of the relation, including the creation of publication-quality plots. Additionally, we study whether the current mass and radius observations of small exoplanets support the presence of a population of water-rich planets positioned between rocky planets and sub-Neptunes. The study is based on Bayesian model comparison and shows somewhat strong support against the existence of a water-world population around M dwarfs. However, the results of the study depend on the chosen theoretical water-world density model. A more conclusive result requires a larger sample of precisely characterised planets and community consensus on a realistic water world interior structure and atmospheric composition model.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352651","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 : 2023-11-13DOI: 10.48550/arxiv.2311.07426
Chan, Alex J., Huyuk, Alihan, van der Schaar, Mihaela
Machine learning models are being increasingly deployed to take, or assist in taking, complicated and high-impact decisions, from quasi-autonomous vehicles to clinical decision support systems. This poses challenges, particularly when models have hard-to-detect failure modes and are able to take actions without oversight. In order to handle this challenge, we propose a method for a collaborative system that remains safe by having a human ultimately making decisions, while giving the model the best opportunity to convince and debate them with interpretable explanations. However, the most helpful explanation varies among individuals and may be inconsistent across stated preferences. To this end we develop an algorithm, Ardent, to efficiently learn a ranking through interaction and best assist humans complete a task. By utilising a collaborative approach, we can ensure safety and improve performance while addressing transparency and accountability concerns. Ardent enables efficient and effective decision-making by adapting to individual preferences for explanations, which we validate through extensive simulations alongside a user study involving a challenging image classification task, demonstrating consistent improvement over competing systems.
{"title":"Optimising Human-AI Collaboration by Learning Convincing Explanations","authors":"Chan, Alex J., Huyuk, Alihan, van der Schaar, Mihaela","doi":"10.48550/arxiv.2311.07426","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07426","url":null,"abstract":"Machine learning models are being increasingly deployed to take, or assist in taking, complicated and high-impact decisions, from quasi-autonomous vehicles to clinical decision support systems. This poses challenges, particularly when models have hard-to-detect failure modes and are able to take actions without oversight. In order to handle this challenge, we propose a method for a collaborative system that remains safe by having a human ultimately making decisions, while giving the model the best opportunity to convince and debate them with interpretable explanations. However, the most helpful explanation varies among individuals and may be inconsistent across stated preferences. To this end we develop an algorithm, Ardent, to efficiently learn a ranking through interaction and best assist humans complete a task. By utilising a collaborative approach, we can ensure safety and improve performance while addressing transparency and accountability concerns. Ardent enables efficient and effective decision-making by adapting to individual preferences for explanations, which we validate through extensive simulations alongside a user study involving a challenging image classification task, demonstrating consistent improvement over competing systems.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352673","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 : 2023-11-13DOI: 10.48550/arxiv.2311.07293
Sekine, Akihiko, Ohfuchi, Mari, Doi, Yoshiyasu
The quantum transduction between microwave and optical photons is essential for realizing scalable quantum computers with superconducting qubits. Due to the large frequency difference between microwave and optical ranges, the transduction needs to be done via intermediate bosonic modes or nonlinear processes. So far, the transduction efficiency $eta$ via the magneto-optic Faraday effect (i.e., the light-magnon interaction) in the ferromagnet YIG has been demonstrated to be small as $etasim 10^{-8} mathrm{-} 10^{-15}$ due to the sample size limitation inside the cavity. Here, we take advantage of the fact that three-dimensional topological insulator thin films exhibit a topological Faraday effect that is independent of the sample thickness. This leads to a large Faraday rotation angle and therefore enhanced light-magnon interaction in the thin film limit. We show theoretically that the transduction efficiency can be greatly improved to $etasim10^{-4}$ by utilizing the heterostructures consisting of topological insulator thin films such as Bi$_2$Se$_3$ and ferromagnetic insulator thin films such as YIG.
{"title":"Microwave-to-Optical Quantum Transduction Utilizing the Topological\u0000 Faraday Effect of Topological Insulator Heterostructures","authors":"Sekine, Akihiko, Ohfuchi, Mari, Doi, Yoshiyasu","doi":"10.48550/arxiv.2311.07293","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07293","url":null,"abstract":"The quantum transduction between microwave and optical photons is essential for realizing scalable quantum computers with superconducting qubits. Due to the large frequency difference between microwave and optical ranges, the transduction needs to be done via intermediate bosonic modes or nonlinear processes. So far, the transduction efficiency $eta$ via the magneto-optic Faraday effect (i.e., the light-magnon interaction) in the ferromagnet YIG has been demonstrated to be small as $etasim 10^{-8} mathrm{-} 10^{-15}$ due to the sample size limitation inside the cavity. Here, we take advantage of the fact that three-dimensional topological insulator thin films exhibit a topological Faraday effect that is independent of the sample thickness. This leads to a large Faraday rotation angle and therefore enhanced light-magnon interaction in the thin film limit. We show theoretically that the transduction efficiency can be greatly improved to $etasim10^{-4}$ by utilizing the heterostructures consisting of topological insulator thin films such as Bi$_2$Se$_3$ and ferromagnetic insulator thin films such as YIG.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352797","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 : 2023-11-13DOI: 10.48550/arxiv.2311.07380
Namekata, Kosuke, Airapetian, Vladimir S., Petit, Pascal, Maehara, Hiroyuki, Ikuta, Kai, Inoue, Shun, Notsu, Yuta, Paudel, Rishi R., Arzoumanian, Zaven, Avramova-Boncheva, Antoaneta A., Gendreau, Keith, Jeffers, Sandra V., Marsden, Stephen, Morin, Julien, Neiner, Coralie, Vidotto, Aline A., Shibata, Kazunari
Young solar-type stars frequently produce superflares, serving as a unique window into the young Sun-Earth environments. Large solar flares are closely linked to coronal mass ejections (CMEs) associated with filament/prominence eruptions, but its observational evidence for stellar superflares remains scarce. Here, we present a 12-day multi-wavelength campaign observation of young solar-type star EK Draconis (G1.5V, 50-120 Myr age) utilizing TESS, NICER, and Seimei telescope. The star has previously exhibited blueshifted H$alpha$ absorptions as evidence for a filament eruption associated with a superflare. Our simultaneous optical and X-ray observations identified three superflares of $1.5times10^{33}$ -- $1.2times10^{34}$ erg. We report the first discovery of two prominence eruptions on a solar-type star, observed as blueshifted H$alpha$ emissions at speed of 690 and 430 km s$^{-1}$ and masses of $1.1times10^{19}$ and $3.2times10^{17}$ g, respectively. The faster, massive event shows a candidate of post-flare X-ray dimming with the amplitude of up to $sim$10 %. Several observational aspects consistently point to the occurrence of a fast CME associated with this event. The comparative analysis of the estimated length scales of flare loops, prominences, possible dimming region, and starspots provides the overall picture of the eruptive phenomena. Furthermore, the energy partition of the observed superflares in the optical and X-ray bands is consistent with flares from the Sun, M-dwarfs, and close binaries, yielding the unified empirical relations. These discoveries provide profound implications of impact of these eruptive events on the early Venus, Earth and Mars and young exoplanets.
{"title":"Multi-wavelength Campaign Observations of a Young Solar-type Star, EK\u0000 Draconis I. Discovery of Prominence Eruptions Associated with Superflares","authors":"Namekata, Kosuke, Airapetian, Vladimir S., Petit, Pascal, Maehara, Hiroyuki, Ikuta, Kai, Inoue, Shun, Notsu, Yuta, Paudel, Rishi R., Arzoumanian, Zaven, Avramova-Boncheva, Antoaneta A., Gendreau, Keith, Jeffers, Sandra V., Marsden, Stephen, Morin, Julien, Neiner, Coralie, Vidotto, Aline A., Shibata, Kazunari","doi":"10.48550/arxiv.2311.07380","DOIUrl":"https://doi.org/10.48550/arxiv.2311.07380","url":null,"abstract":"Young solar-type stars frequently produce superflares, serving as a unique window into the young Sun-Earth environments. Large solar flares are closely linked to coronal mass ejections (CMEs) associated with filament/prominence eruptions, but its observational evidence for stellar superflares remains scarce. Here, we present a 12-day multi-wavelength campaign observation of young solar-type star EK Draconis (G1.5V, 50-120 Myr age) utilizing TESS, NICER, and Seimei telescope. The star has previously exhibited blueshifted H$alpha$ absorptions as evidence for a filament eruption associated with a superflare. Our simultaneous optical and X-ray observations identified three superflares of $1.5times10^{33}$ -- $1.2times10^{34}$ erg. We report the first discovery of two prominence eruptions on a solar-type star, observed as blueshifted H$alpha$ emissions at speed of 690 and 430 km s$^{-1}$ and masses of $1.1times10^{19}$ and $3.2times10^{17}$ g, respectively. The faster, massive event shows a candidate of post-flare X-ray dimming with the amplitude of up to $sim$10 %. Several observational aspects consistently point to the occurrence of a fast CME associated with this event. The comparative analysis of the estimated length scales of flare loops, prominences, possible dimming region, and starspots provides the overall picture of the eruptive phenomena. Furthermore, the energy partition of the observed superflares in the optical and X-ray bands is consistent with flares from the Sun, M-dwarfs, and close binaries, yielding the unified empirical relations. These discoveries provide profound implications of impact of these eruptive events on the early Venus, Earth and Mars and young exoplanets.","PeriodicalId":496270,"journal":{"name":"arXiv (Cornell University)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352812","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：