Neuro Causal and Symbolic AI. 36th NIPS
Neuro Causal and Symbolic AI. 36th NIPS
理解因果互动是人类认知的核心，因此也是科学、工程、商业和法律的核心追求。发展心理学表明，儿童探索世界的方式与科学家相似，他们会问诸如“如果”这样的问题以及“为什么？”人工智能研究的目标是在机器中复制这些能力。特别是深度学习通过端到端可训练的深度神经网络，为函数逼近带来了强大的工具。这种能力已经被无数应用中的巨大成功所证实。然而，它们缺乏解释能力和推理能力，这被证明是建立类人能力系统的一个障碍。因此，在深度学习中启用因果推理能力对于研究通向人类智能的道路至关重要。神经因果模型的第一步已经存在，并有望实现人工智能系统的愿景，即像现代神经模型一样高效地进行因果推理。类似地，经典的符号方法正在被重新访问和重新整合到当前的系统中，以允许超越纯模式识别的推理能力。因果关系的Pearlian形式化揭示了一个理论上合理和实践上严格的推理层次，作为评估神经符号系统推理能力的有益基准。我们的目标是将对人工智能研究领域(一般机器和深度学习、符号和以对象为中心的方法以及逻辑)的整合感兴趣的研究人员聚集在一起，并以开发下一代人工智能系统为目标，对因果关系进行严格的形式化。
Accepted (Contributed Talks)
Title | Authors |
|---|---|
GlanceNets: Interpretable, Leak-proof Concept-based Models | Emanuele Marconato, Andrea Passerini, Stefano Teso |
Interventional Causal Representation Learning | Kartik Ahuja, Yixin Wang, Divyat Mahajan, Yoshua Bengio |
Meaning without reference in large language models | Steven Piantadosi, Felix Hill |
Unlocking Slot Attention by Changing Optimal Transport Costs | Yan Zhang, David W Zhang, Simon Lacoste-Julien, Gertjan J. Burghouts, Cees G. M. Snoek |
Accepted (Poster)
Title | Authors |
|---|---|
The Impact of Symbolic Representations on In-context Learning for Few-shot Reasoning | Hanlin Zhang, YiFan Zhang, Li Erran Li, Eric Xing |
Graphs, Constraints, and Search for the Abstraction and Reasoning Corpus | Yudong Xu, Elias Boutros Khalil, Scott Sanner |
Counterfactual reasoning: Do Language Models need world knowledge for causal inference? | Jiaxuan Li, Lang Yu, Allyson Ettinger |
Image Manipulation via Neuro-Symbolic Networks | Harman Singh, Poorva Garg, Mohit Gupta, Kevin Shah, Arnab Kumar Mondal, Dinesh Khandelwal, Parag Singla, Dinesh Garg |
Learning Neuro-symbolic Programs for Language-Guided Robotic Manipulation | Namasivayam Kalithasan, Himanshu Gaurav Singh, Vishal Bindal, Arnav Tuli, Vishwajeet Agrawal, Rahul Jain, Parag Singla, Rohan Paul |
Active Bayesian Causal Inference | Christian Toth, Lars Lorch, Christian Knoll, Andreas Krause, Franz Pernkopf, Robert Peharz, Julius Von Kügelgen |
Causal Discovery for Modular World Models | Anson Lei, Bernhard Schölkopf, Ingmar Posner |
Playgrounds for Abstraction and Reasoning | Subin Kim, Prin Phunyaphibarn, Donghyun Ahn, Sundong Kim |
Enhancing Transfer of Reinforcement Learning Agents with Abstract Contextual Embeddings | Guy Azran, Mohamad Hosein Danesh, Stefano V Albrecht, Sarah Keren |
Hierarchical Abstraction for Combinatorial Generalization in Object Rearrangement | Michael Chang, Alyssa Li Dayan, Franziska Meier, Thomas L. Griffiths, Sergey Levine, Amy Zhan |
Benchmarking Counterfactual Reasoning Abilities about Implicit Physical Properties | Maitreya Patel, Tejas Gokhale, Chitta Baral, Yezhou Yang |
Symbolic Causal Inference via Operations on Probabilistic Circuits | Benjie Wang, Marta Kwiatkowska |
Discrete Learning Of DAGs Via Backpropagation | Andrew J. Wren, Pasquale Minervini, Luca Franceschi, Valentina Zantedeschi |
Probabilities of Causation: Adequate Size of Experimental and Observational Samples | Ang Li, Ruirui Mao, Judea Pearl |
Synthesized Differentiable Programs | Lucas Paul Saldyt |
其他参考：