https://www.brianzhou.org/coauthors/brian-zhou Research publications and working papers coauthored with Brian Zhou. en-us hello@zhoubrian.com (Brian Zhou) hello@zhoubrian.com (Brian Zhou) Wed, 01 Jan 2025 00:00:00 GMT https://www.hks.harvard.edu/centers/mrcbg/publications/awp/awp251#governance-at-a-crossroads-artificial-intelligence-and-the-future-of-innovation-in-america-ssrn https://www.hks.harvard.edu/centers/mrcbg/publications/awp/awp251 We present ACO-ToT, an algorithm that leverages fine-tuned LLM "ants" guided by ant colony optimization to efficiently uncover optimal reasoning paths for complex problems. Wed, 01 Jan 2025 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsSSRN https://arxiv.org/abs/2501.19278#pheromone-based-learning-of-optimal-reasoning-paths-arxiv-under-review https://arxiv.org/abs/2501.19278 We present ACO-ToT, an algorithm that leverages fine-tuned LLM "ants" guided by ant colony optimization to efficiently uncover optimal reasoning paths for complex problems. Wed, 01 Jan 2025 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsarXiv (under review) https://arxiv.org/abs/2501.19318#mindstores-memory-informed-neural-decision-synthesis-for-task-oriented-reinforcement-in-embodied-systems-iclr-workshop-on-reasoning-and-planning-for-llms-arxiv-under-review https://arxiv.org/abs/2501.19318 We present MINDSTORES, an experience-augmented planning framework that enables embodied agents to build and leverage mental models through natural interaction, improving zero-shot LLM planning for complex open-world tasks. Wed, 01 Jan 2025 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsICLR Workshop on Reasoning and Planning for LLMs; arXiv (under review) https://www.mdpi.com/2313-7673/9/7/434#insights-into-flexible-bioinspired-fins-for-unmanned-underwater-vehicle-systems-through-deep-learning-neurips-workshop-on-machine-learning-and-the-physical-sciences https://www.mdpi.com/2313-7673/9/7/434 We introduce forward neural network models that integrate fin stiffness to predict UUV kinematics, thrust, and efficiency, improving performance in bio-inspired fin designs. Mon, 01 Jan 2024 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsNeurIPS Workshop on Machine Learning and the Physical Sciences https://www.mdpi.com/2313-7673/9/7/434#insights-into-flexible-bioinspired-fins-for-unmanned-underwater-vehicle-systems-through-deep-learning-biomimetics https://www.mdpi.com/2313-7673/9/7/434 We introduce forward neural network models that integrate fin stiffness to predict UUV kinematics, thrust, and efficiency, improving performance in bio-inspired fin designs. Mon, 01 Jan 2024 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsBiomimetics https://aapor.confex.com/aapor/2024/meetingapp.cgi/Paper/3181#a-framework-to-apply-natural-language-processing-techniques-to-analyze-public-opinions-on-peace-and-governance-in-africa-aapor-conference https://aapor.confex.com/aapor/2024/meetingapp.cgi/Paper/3181 We aggregate corpora from UN General Debates and emerging media for conflict prediction. Mon, 01 Jan 2024 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsAAPOR Conference https://aapor.confex.com/aapor/2024/meetingapp.cgi/Paper/2503#analyzing-post-covid-19-changes-to-voter-behavior-and-early-voting-trends-in-virginia-aapor-conference https://aapor.confex.com/aapor/2024/meetingapp.cgi/Paper/2503 We analyze changes in early voting and absentee voting rates in Virginia since the 2020 General Election, examining factors like voter enthusiasm, party initiatives, and political issues to understand post-COVID voting behavior. Mon, 01 Jan 2024 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsAAPOR Conference https://ojs.aaai.org/index.php/AAAI/article/view/30538#power-aware-inverse-search-machine-learning-for-low-resource-multi-objective-unmanned-underwater-vehicle-control-aaai-conference-on-artificial-intelligence https://ojs.aaai.org/index.php/AAAI/article/view/30538 Developed power-aware machine learning models for underewater vehicle control, improving efficiency while maintaining performance. Mon, 01 Jan 2024 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsAAAI Conference on Artificial Intelligence https://www.brianzhou.org/static/research/posters/2024_qlab_agents.pdf#development-of-quantum-machine-learning-agents-to-model-simple-economies-working-paper https://www.brianzhou.org/static/research/posters/2024_qlab_agents.pdf We propose a new approach to agent-based modeling that integrates quantum computing, creating adaptive learning agents to simulate decision-making in complex environments, offering performance benefits over traditional models. Mon, 01 Jan 2024 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) Working PapersWorking Paper https://ieeexplore.ieee.org/abstract/document/10534975#analyzing-the-discourse-in-the-un-for-crisis-response-in-post-colonial-africa-ieee-mit-urtc https://ieeexplore.ieee.org/abstract/document/10534975 We analyze the UN General Debate Corpus to track the influence of African post-colonial states, using natural language processing to forecast shifts in global priorities and the efficacy of crisis resolution measures. Sun, 01 Jan 2023 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsIEEE MIT URTC https://ojs.aaai.org/index.php/AAAI/article/view/26863#data-driven-machine-learning-models-for-a-multi-objective-flapping-fin-unmanned-underwater-vehicle-control-system-aaai-conference-on-artificial-intelligence https://ojs.aaai.org/index.php/AAAI/article/view/26863 We develop a search-based inverse model that leverages a kinematics-to-thrust neural network to determine flapping fin kinematics achieving target thrust and smooth transitions, enabling real-time control adjustments for UUV propulsion. Sun, 01 Jan 2023 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsAAAI Conference on Artificial Intelligence https://blog.genlaw.org/papers.html#licensing-training-data-and-attributing-copyright-of-derivative-content-from-large-language-models-can-resolve-up--and-downstream-copyright-issues#licensing-training-data-and-attributing-copyright-of-derivative-content-from-large-language-models-can-resolve-up--and-downstream-copyright-issues-icml-workshop-on-generative-ai-and-the-law https://blog.genlaw.org/papers.html#licensing-training-data-and-attributing-copyright-of-derivative-content-from-large-language-models-can-resolve-up--and-downstream-copyright-issues We propose an opt-in system with multimodal similarity measures and metadata tagging for fair IP compensation and attribution, resolving copyright disputes over LLM-generated content. Sun, 01 Jan 2023 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsICML Workshop on Generative AI and the Law https://arxiv.org/abs/2310.14135#computational-approaches-for-modeling-power-consumption-on-an-underwater-flapping-fin-propulsion-system-aaai-symposium-on-knowledge-guided-machine-learning https://arxiv.org/abs/2310.14135 We develop a non-dimensional figure of merit (FOM) to evaluate fin designs and kinematics for bio-inspired UUVs, using computational models to predict thrust, power, and efficiency, optimizing gait performance and material selection. Sat, 01 Jan 2022 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsAAAI Symposium on Knowledge-Guided Machine Learning https://osf.io/preprints/5jkum#identification-of-ocular-biomarkers-for-the-development-of-an-early-stage-diagnostic-tool-for-neurodegenerative-disease-osf-preprints https://osf.io/preprints/5jkum We identify ocular biomarkers and define data collection procedures for saccade and pursuit tasks to develop a gaze-based diagnostic tool for early neurodegenerative disease detection. Sat, 01 Jan 2022 00:00:00 GMT hello@zhoubrian.com (Brian Zhou) PublicationsOSF Preprints