The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence

2020/02/14 by Gary Marcus, Marcus, Gary · 9 voices · 36 citations
Computer Science · #cs.AI #cs.LG

paper · pdf · doi:10.48550/arxiv.2002.06177

Abstract

Recent research in artificial intelligence and machine learning has largely emphasized general-purpose learning and ever-larger training sets and more and more compute. In contrast, I propose a hybrid, knowledge-driven, reasoning-based approach, centered around cognitive models, that could provide the substrate for a richer, more robust AI than is currently possible.

Citations

• Compositional Semantic Parsing on Semi-Structured Tables
• Deep Learning for Symbolic Mathematics
• Environmental drivers of systematicity and generalization in a situated agent

Cited by

• Confidently Wrong: Exception Chain Collapse in Frontier LLM Rule Evaluation
• Beyond Context: Large Language Models Failure to Grasp Users Intent
• From artificial to organic: Rethinking the roots of intelligence for digital health
• Neurosymbolic Inference On Foundation Models For Remote Sensing Text-to-image Retrieval With Complex Queries
• Developmental Symmetry-Loss: A Free-Energy Perspective on Brain-Inspired Invariance Learning
• Addressing Logical Fallacies In Scientific Reasoning From Large Language Models: Towards a Dual-Inference Training Framework
• Probabilistic Neuro-Symbolic Reasoning for Sparse Historical Data: A Framework Integrating Bayesian Inference, Causal Models, and Game-Theoretic Allocation
• Distilling Counterfactual Reasoning from Language to Vision: Causal Graph Guided Post-Training for Video Understanding
• Decrypting Cryptic Crosswords: Semantically Complex Wordplay Puzzles as a Target for NLP
• Dynamic Stability of LLM-Generated Code
• Learning Algebraic Representation for Systematic Generalization in Abstract Reasoning
• Graph Neural Networks Meet Neural-Symbolic Computing: A Survey and\n Perspective
• Spectral imaginings and sympoietic creativity: AI hallucinations and the ethics of posthuman creativity
• The Future of Human Responsibility
• Retrieval-Augmented Generation-aided causal identification of aviation accidents: A large language model methodology
• A Review of Winograd Schema Challenge Datasets and Approaches
• Robust Conversational AI with Grounded Text Generation
• LLM-Augmented Symbolic NLU System for More Reliable Continuous Causal Statement Interpretation
• ActivationReasoning: Logical Reasoning in Latent Activation Spaces
• Towards Neurocognitive-Inspired Intelligence: From AI's Structural Mimicry to Human-Like Functional Cognition
• MIMIC: Integrating Diverse Personality Traits for Better Game Testing Using Large Language Model
• A survey on neural-symbolic learning systems
• Artificial Intelligence’s new clothes? A system technology perspective
• How Can We Characterize Human Generalization and Distinguish It From Generalization in Machines?
• TERAG: Token-Efficient Graph-Based Retrieval-Augmented Generation
• Neuromorphic Intelligence
• Out of One, Many: Using Language Models to Simulate Human Samples
• Discovering Symbolic Models from Deep Learning with Inductive Biases
• An Approach to Grounding AI Model Evaluations in Human-derived Criteria
• A Comprehensive Review of AI Agents: Transforming Possibilities in Technology and Beyond
• UNVEILING: What Makes Linguistics Olympiad Puzzles Tricky for LLMs?
• Understanding in Artificial Intelligence
• PLATO-XL: Exploring the Large-scale Pre-training of Dialogue Generation
• Quantum Spectral Reasoning: A Non-Neural Architecture for Interpretable Machine Learning
• Categorical Construction of Logically Verifiable Neural Architectures
• Applying Psychometrics to Large Language Model Simulated Populations: Recreating the HEXACO Personality Inventory Experiment with Generative Agents

Discussions

• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence [hn, 6 points, 0 comments]
• The Next Decade in AI (2020) [hn, 3 points, 0 comments]
• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence [hn, 3 points, 1 comments]
• Thanks interesting read! I enjoyed @garymarcus.bsky.social's "Next Decade In AI Paper" which sounds similar to your article? It's all interesting, but as I understand, still largely speculative and u [bsky, 3 points, 0 comments]
• Four Steps Towards Robust Artificial Intelligence (2020) [hn, 2 points, 1 comments]
• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence [hn, 2 points, 0 comments]
• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence [hn, 2 points, 0 comments]
• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence [hn, 1 points, 0 comments]
• The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence [hn, 1 points, 0 comments]

Related

• Making AI 'Smart': Bridging AI and Cognitive Science
• Artificial Intelligence and its Role in Near Future
• Next Wave Artificial Intelligence: Robust, Explainable, Adaptable, Ethical, and Accountable
• World Models in Artificial Intelligence: Sensing, Learning, and Reasoning Like a Child
• Artificial Intelligence in the Rising Wave of Deep Learning: The Historical Path and Future Outlook [Perspectives]