Scribd
Upload
1 views17 pages

AI Problem Map

The document outlines the AI Problem Map, detailing the goals of AI, which include perception, reasoning, planning, and action across various data types such as text, images, audio, video, tabular data, and graphs. It discusses core tasks and current trends in each data type, emphasizing the importance of understanding how data shapes AI capabilities. The conclusion suggests starting with a small personal project as a practical approach to applying AI concepts.

Uploaded by

haquynhanh1106
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
1 views17 pages

AI Problem Map

The document outlines the AI Problem Map, detailing the goals of AI, which include perception, reasoning, planning, and action across various data types such as text, images, audio, video, tabular data, and graphs. It discusses core tasks and current trends in each data type, emphasizing the importance of understanding how data shapes AI capabilities. The conclusion suggests starting with a small personal project as a practical approach to applying AI concepts.

Uploaded by

haquynhanh1106
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

NATIONAL ECONOMICS UNIVERSITY

FACULTY OF INFORMATION TECHNOLOGY

AI Problem Map by Data Type

Nguyen Quoc Thai


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

What Is AI and What Are Its Goals?


AI (practical definition): systems that can perceive, understand/reason, plan/decide, and sometimes act

Three goal layers


Capability: Perception → Reasoning →
Planning → Action
Engineering: quality, generalization,
trust, efficiency
Product: automation, productivity,
personalization, new capabilities
AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

AI Across Data Types


Each data type shapes what “understanding” means:

Text: meaning, intent, facts, structure


Image: objects, scenes, documents
Audio: speech + speaker + acoustic cues
Video: events over time
(spatiotemporal)
Tabular: feature-driven prediction with
business constraints
Graph/3D/Robotics: relations, space,
interaction, control
AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Text (NLP) - Core Task


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Text (NLP) - Current Trends


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Image (Vision) - Core Task


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Image (Vision) - Current Trends


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Audio/Speech - Core Task


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Audio/Speech - Current Trends


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Video - Core Task


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Video - Current Trends


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Tabular/Tables - Core Task


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Tabular/Tables - Current Trends


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Graph AI: Core Tasks & Current Trends


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

From Core Task to Real-World Use Cases


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

My Current Research Directions


AI Problem Map INTRODUCTION TASKS MY RESEARCH CONCLUSION

Conclusion
No Wrong Choices, Only Right Fits

A practical recommendation: begin with a small personal project

Nguyen Quoc Thai - quocthai@[Link]

You might also like