Published Feb 8, 2024

Subbarao Kambhampati: Planning, Reasoning, and Interpretability in the Age of LLMs

Daniel Bashir converses with Subbarao Kambhampati about the complex interplay of tacit and symbolic knowledge in AI, tackling the challenges of planning, reasoning, and interpretability amid the rise of large language models (LLMs) and the necessity for effective human-AI interactions.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Cameron Jones & Sean Trott: Understanding, Grounding, and Reference in LLMs
Answers 383 questions
Been Kim: Interpretable Machine Learning
Answers 383 questions
Harvey Lederman: Propositional Attitudes and Reference in Language Models
Answers 383 questions
Stevan Harnad: AI's Symbol Grounding Problem
Answers 383 questions
Melanie Mitchell: Abstraction and Analogy in AI
Answers 383 questions
Venkatesh Rao: Protocols, Intelligence, and Scaling
Answers 383 questions
Percy Liang on Machine Learning Robustness, Foundation Models, and Reproducibility
Answers 383 questions
Arjun Ramani & Zhengdong Wang: Why Transformative AI is Really, Really Hard to Achieve
Answers 383 questions
Catherine Olsson and Nelson Elhage: Anthropic, Understanding Transformers
Answers 383 questions
Manuel & Lenore Blum: The Conscious Turing Machine
Answers 383 questions
Terry Winograd: AI, HCI, Language, and Cognition
Answers 383 questions
Stuart Russell: The Foundations of Artificial Intelligence
Answers 383 questions
Vivek Natarajan: Towards Biomedical AI
Answers 383 questions
Jacob Andreas: Language, Grounding, and World Models
Answers 383 questions
Peter Lee: Computing Theory and Practice, and GPT-4's Impact
Answers 383 questions

Dexa/Daniel Bashir

Subbarao Kambhampati: Planning, Reasoning, and Interpretability in the Age of LLMs

Topics covered

Popular Clips

Teasing Out Symbolic Knowledge

Emergent AI Capabilities

Language and Thought

AI Journey Origins

Language Models' Interpretability

AI Reasoning Architectures

Computational Complexity Insights

AI Plan Critiquing

AI Knowledge Tasks

AI Assistants' Effectiveness

Neurosymbolic Abstractions

System Two Integration

Evolution of AI

Unveiling Machine Understanding

Episode Highlights

Symbolic and Tacit Knowledge

Planning and Reasoning

Explanations and Interpretability

Related Episodes

Cameron Jones & Sean Trott: Understanding, Grounding, and Reference in LLMs

Been Kim: Interpretable Machine Learning

Harvey Lederman: Propositional Attitudes and Reference in Language Models

Stevan Harnad: AI's Symbol Grounding Problem

Melanie Mitchell: Abstraction and Analogy in AI

Venkatesh Rao: Protocols, Intelligence, and Scaling

Percy Liang on Machine Learning Robustness, Foundation Models, and Reproducibility

Arjun Ramani & Zhengdong Wang: Why Transformative AI is Really, Really Hard to Achieve

Catherine Olsson and Nelson Elhage: Anthropic, Understanding Transformers

Manuel & Lenore Blum: The Conscious Turing Machine

Terry Winograd: AI, HCI, Language, and Cognition

Stuart Russell: The Foundations of Artificial Intelligence

Vivek Natarajan: Towards Biomedical AI

Jacob Andreas: Language, Grounding, and World Models

Peter Lee: Computing Theory and Practice, and GPT-4's Impact

Subbarao Kambhampati: Planning, Reasoning, and Interpretability in the Age of LLMs

Topics covered

Popular Clips

Episode Highlights

Symbolic and Tacit KnowledgeSubbarao Kambhampati examines the intricate dynamics of tacit and explicit knowledge in AI systems. He discusses how symbolic reasoning and human-AI knowledge exchange shape the development of new concepts and improve interaction.

Symbolic and Tacit Knowledge

Planning and ReasoningSubbarao Kambhampati explores the intricate challenges of planning complexity in AI, emphasizing the role of neurosymbolic approaches and human collaboration. He highlights the limitations of LLMs in reasoning and the importance of external verification in AI planning.

Planning and Reasoning

Explanations and InterpretabilitySubbarao Kambhampati explores the importance of explanations in AI, highlighting challenges in interpretability and effective communication. He discusses how AI systems can bridge the gap between machine reasoning and human understanding through symbolic models.

Explanations and Interpretability

Related Episodes