我才博一,我得想想我能做点什么? 导师在年初3月份的时候组织过一波具身智能(Embodied AI)的交流会。EMBODIED AI 最初由阿兰·图灵在1950年提出的具身图灵测试(Embodied Turing Test)中提出,该测试旨在确定代理是否能够展示出不仅限于解决虚拟环境(网络空间)中的抽象问题的智能,还能够应对物理世界的复杂性和不可预测性...
Calvaresi D, Ciatto G, Najjar A, Aydoğan R, Torre L, Omicini A, Schumacher M (2021) Expectation: personalized explainable artificial intelligence for decentralized agents with heterogeneous knowledge. In: Explainable and transparent AI and multi-agent systems: third international workshop, EXTRAAMA...
可解释的AI(XAI)生成细节或理由,使AI系统功能清晰或易于理解。为了使深度学习具有可解释性,一些工作提出了视觉解释技术,如归因或显著性图。但这些解释并不总是足够的,不容易量化、修正,也不容易给非技术用户传达。 要部署以人为本的AI系统,就要扩大不同的少数群体和受众的包容,这可以促进对AI模型的评估。 因此深...
Towards integrated neural-symbolic systems for human-level AI: Two research programs helping to bridge the gaps. Biologically Inspired Cognitive Architectures, 14: 97-110, 2015Besold, T.R., Ku¨hnberger, K.U.: Towards integrated neural-symbolic systems for human-level AI: Two research programs...
Over the last decades, deep neural networks based-models became the dominant paradigm in machine learning. Further, the use of artificial neural networks in symbolic learning has been seen as increasingly relevant recently. To study the capabilities of neural networks in the symbolic AI domain, rese...
An overview on the current state of AI-driven battery research, which until recently was hampered by a lack of data but is now poised to pick up the pace — Improving batteries has always been hampered by slow experimentation and discovery processes. Machine learning is speeding it up by or...
A joint venture between IBM and MIT relies on machine vision in combination with a broad AI that can multitask and cover multiple domains, but which also can read structured and unstructured data from a variety of sources
Here we use Monte Carlo tree search and symbolic artificial intelligence (AI) to discover retrosynthetic routes. We combined Monte Carlo tree search with an expansion policy network that guides the search, and a filter network to pre-select the most promising retrosynthetic steps. These deep neural...
Neural networks for video game AI. In Artificial and Computational Intelligence in Games: Integration (Dagstuhl Seminar 15051) Vol. 5 (eds Lucas, S. M. et al.) 224 (2015) Silver, D. et al. Mastering the game of Go with deep neural networks and tree search. Nature 529, 484–489 (...
内容提示: Neural-Symbolic VQA: Disentangling Reasoningfrom Vision and Language UnderstandingKexin Yi ∗Harvard UniversityJiajun Wu ∗MIT CSAILChuang GanMIT-IBM Watson AI LabAntonio TorralbaMIT CSAILPushmeet KohliDeepMindJoshua B. TenenbaumMIT CSAILAbstractWe marry two powerful ideas: deep representation ...