Dexa/Machine Learning Street Talk (MLST)

Published Dec 1, 2024

Jonas Hübotter (ETH) - Test Time Inference

PhD student Jonas Hübotter from ETH Zurich discusses groundbreaking advancements in AI test-time computation, emphasizing resource optimization, adaptive systems, and the innovative use of smaller models to outperform larger ones. The episode explores hybrid deployment strategies, local learning methods, and the evolution of information retrieval, challenging traditional machine learning paradigms and enhancing decision-making capabilities.

Episode Highlights

Topics covered

Questions from this episode

- Will large language models scale all the way to artificial general intelligence (AGI)?
  Asked by 143 people
- Tell me something about artificial intelligence
  Asked by 121 people
- Is artificial intelligence really intelligent?
  Asked by 102 people
- What can we expect in the future of AI?
  Asked by 72 people
- Can you give more examples of large language model applications in relation to the episode Stephen Wolfram: ChatGPT and the Nature of Truth, Reality & Computation | Lex Fridman Podcast #376 and the clip Computational Language Insights?
  Asked by 50 people
- How does AI transform the learning process?
  Asked by 48 people
- What's the future of generative AI?
  Asked by 42 people
- How is machine learning evolving?
  Asked by 32 people
- Is reinforcement learning a turning point for large language models (LLMs) and artificial intelligence (AI)?
  Asked by 29 people
- What are the most recent developments and trends in AI, large language models (LLMs), retrieval-augmented generation (RAG), etc., as discussed in the episode #159 - Inflection-2.5, Devin, OpenAI board update, SIMA, EU AI Act passed and the clip Cohere's Command R?
  Asked by 28 people
- Can AI systems be scaled?
  Asked by 27 people
- What techniques are used with large language models (LLMs)?
  Asked by 23 people
- How do you leverage different models in machine learning?
  Asked by 20 people
- Tell me something unique about large language models
  Asked by 20 people
- What are the potential advancements in artificial intelligence?
  Asked by 17 people

Episode Highlights

Related Episodes

Dr. Sanjeev Namjoshi - Active Inference
Answers 383 questions
ICLR 2020: Yoshua Bengio and the Nature of Consciousness
Answers 383 questions
Prof. BERT DE VRIES - ON ACTIVE INFERENCE
Answers 383 questions
Dr. Thomas Parr - Active Inference Book
Answers 383 questions
Jürgen Schmidhuber - Neural and Non-Neural AI, Reasoning, Transformers, and LSTMs
Answers 383 questions
Speechmatics CTO - Next-Generation Speech Recognition
Answers 383 questions
Mahault Albarracin - Cognitive Science
Answers 383 questions
DR. JEFF BECK - THE BAYESIAN BRAIN
Answers 383 questions
Computation, Bayesian Model Selection, Interactive Articles
Answers 383 questions
Sepp Hochreiter - LSTM: The Comeback Story?
Answers 383 questions
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Answers 383 questions
#82 - Dr. JOSCHA BACH - Digital Physics, DL and Consciousness [UNPLUGGED]
Answers 383 questions
Pattern Recognition vs True Intelligence - Francois Chollet
Answers 383 questions
WelcomeAIOverlords (Zak Jost)
Answers 383 questions
#65 Prof. PEDRO DOMINGOS [Unplugged]
Answers 383 questions