Concept Learning with Energy-Based Models (Paper Explained)
This is a hard paper! Energy-functions are typically a mere afterthought in current machine learning. A core function of the Energy - its smoothness - is usually not exploited at inference time. This paper takes a stab at it. Inferring concepts, world states, and attention masks via gradient descent on a learned energy function leads to an interesting framework with many possibilities.
Paper: https://arxiv.org/abs/1811.02486
Blog: https://openai.com/blog/learning-concepts-with-energy-functions/
Videos: https://sites.google.com/site/energyconceptmodels/
Abstract:
Many hallmarks of human intelligence, such as generalizing from limited experience, abstract reasoning and planning, analogical reasoning, creative problem solving, and capacity for language require the ability to consolidate experience into concepts, which act as basic building blocks of understanding and reasoning. We present a framework that defines a concept by an energy function over events in the environment, as well as an attention mask over entities participating in the event. Given few demonstration events, our method uses inference-time optimization procedure to generate events involving similar concepts or identify entities involved in the concept. We evaluate our framework on learning visual, quantitative, relational, temporal concepts from demonstration events in an unsupervised manner. Our approach is able to successfully generate and identify concepts in a few-shot setting and resulting learned concepts can be reused across environments. Example videos of our results are available at this http URL
Authors: Igor Mordatch
Links:
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Видео Concept Learning with Energy-Based Models (Paper Explained) канала Yannic Kilcher
Paper: https://arxiv.org/abs/1811.02486
Blog: https://openai.com/blog/learning-concepts-with-energy-functions/
Videos: https://sites.google.com/site/energyconceptmodels/
Abstract:
Many hallmarks of human intelligence, such as generalizing from limited experience, abstract reasoning and planning, analogical reasoning, creative problem solving, and capacity for language require the ability to consolidate experience into concepts, which act as basic building blocks of understanding and reasoning. We present a framework that defines a concept by an energy function over events in the environment, as well as an attention mask over entities participating in the event. Given few demonstration events, our method uses inference-time optimization procedure to generate events involving similar concepts or identify entities involved in the concept. We evaluate our framework on learning visual, quantitative, relational, temporal concepts from demonstration events in an unsupervised manner. Our approach is able to successfully generate and identify concepts in a few-shot setting and resulting learned concepts can be reused across environments. Example videos of our results are available at this http URL
Authors: Igor Mordatch
Links:
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Видео Concept Learning with Energy-Based Models (Paper Explained) канала Yannic Kilcher
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Energy-based Approaches to Representation Learning - Yann LeCunSupervised Contrastive LearningLearning To Classify Images Without Labels (Paper Explained)[ML News] EU regulates AI, China trains 1.75T model, Google's oopsie, Everybody cheers for fraud.The Rise of Artificial Intelligence through Deep Learning | Yoshua Bengio | TEDxMontrealWhat Is Green Hydrogen And Will It Power The Future?What are Normalizing Flows?CUDA Explained - Why Deep Learning uses GPUsContext R-CNN: Long Term Temporal Context for Per-Camera Object Detection (Paper Explained)[ML News] DeepMind fails to get independence from Google[ML News] Anthropic raises $124M, ML execs clueless, collusion rings, ELIZA source discovered & moreFNet: Mixing Tokens with Fourier Transforms (Machine Learning Research Paper Explained)How much memory does Longformer use?[ML News] De-Biasing GPT-3 | RL cracks chip design | NetHack challenge | Open-Source GPT-JMachine Learning PhD Survival Guide 2021 | Advice on Topic Selection, Papers, Conferences & more![Classic] Word2Vec: Distributed Representations of Words and Phrases and their Compositionality[Classic] Deep Residual Learning for Image Recognition (Paper Explained)SinGAN: Learning a Generative Model from a Single Natural ImageAttention in Neural NetworksGroup Normalization (Paper Explained)