Christopher D Manning: A Neural Network Model That Can Reason (ICLR 2018 invited talk)

Автор: Steven Van Vaerenbergh

Загружено: 2018-05-04

Просмотров: 10284

Описание: Abstract: Deep learning has had enormous success on perceptual tasks but still struggles in providing a model for inference. To address this gap, we have been developing Memory-Attention-Composition networks (MACnets). The MACnet design provides a strong prior for explicitly iterative reasoning, enabling it to learn explainable, structured reasoning, as well as achieve good generalization from a modest amount of data. The model builds from the great success of existing recurrent cells such as LSTMs: A MacNet is a sequence of a single recurrent Memory, Attention, and Composition (MAC) cell. However, its design imposes structural constraints on the operation of each cell and the interactions between them, incorporating explicit control and soft attention mechanisms. We demonstrate the model’s strength and robustness on the challenging CLEVR dataset for visual reasoning (Johnson et al. 2016), achieving a new state-of-the-art 98.9% accuracy, halving the error rate of the previous best model. More importantly, we show that the new model is more data-efficient, achieving good results from even a modest amount of training data. Joint work with Drew Hudson.

Не удается загрузить Youtube-плеер. Проверьте блокировку Youtube в вашей сети.
Повторяем попытку...

Christopher D Manning: A Neural Network Model That Can Reason (ICLR 2018 invited talk)

Доступные форматы для скачивания:

Скачать видео

Информация по загрузке:

Скачать аудио

Похожие видео

Building Neural Network Models That Can Reason

Building Neural Network Models That Can Reason

Kristen Grauman: Visual Learning With Unlabeled Video and Look-Around Policies ICLR2018 invited talk

Kristen Grauman: Visual Learning With Unlabeled Video and Look-Around Policies ICLR2018 invited talk

Neural and Non-Neural AI, Reasoning, Transformers, and LSTMs

Neural and Non-Neural AI, Reasoning, Transformers, and LSTMs

Ian Goodfellow: Adversarial Machine Learning (ICLR 2019 invited talk)

Ian Goodfellow: Adversarial Machine Learning (ICLR 2019 invited talk)

Positive Mood Jazz ☕ Cozy Winter Coffee Jazz Music and Sweet Bossa Nova Piano for Energy the day

Positive Mood Jazz ☕ Cozy Winter Coffee Jazz Music and Sweet Bossa Nova Piano for Energy the day

What If You Keep Slowing Down?

What If You Keep Slowing Down?

Variational Inference: Foundations and Modern Methods (NIPS 2016 tutorial)

Variational Inference: Foundations and Modern Methods (NIPS 2016 tutorial)

Joelle Pineau: Reproducibility, Reusability, and Robustness in Deep Reinforcement Learning ICLR 2018

Joelle Pineau: Reproducibility, Reusability, and Robustness in Deep Reinforcement Learning ICLR 2018

RNN Symposium 2016: Alex Graves - Differentiable Neural Computer

RNN Symposium 2016: Alex Graves - Differentiable Neural Computer

Что происходит с нейросетью во время обучения?

Что происходит с нейросетью во время обучения?

LSTM is dead. Long Live Transformers!

LSTM is dead. Long Live Transformers!

Theoretical Foundations of Graph Neural Networks

Theoretical Foundations of Graph Neural Networks

Geoffrey Hinton: The Foundations of Deep Learning

Geoffrey Hinton: The Foundations of Deep Learning

Rethinking Physics Informed Neural Networks [NeurIPS'21]

Rethinking Physics Informed Neural Networks [NeurIPS'21]

Eric J. Ma - An Attempt At Demystifying Bayesian Deep Learning

Eric J. Ma - An Attempt At Demystifying Bayesian Deep Learning

Nick Shea: The Importance of Logical Reasoning and Its Emergence in Deep Neural Networks

Nick Shea: The Importance of Logical Reasoning and Its Emergence in Deep Neural Networks

Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM)

Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM)

Yann LeCun and Christopher Manning discuss Deep Learning and Innate Priors

Yann LeCun and Christopher Manning discuss Deep Learning and Innate Priors

Bernhard Schölkopf: Learning Causal Mechanisms (ICLR invited talk)

Bernhard Schölkopf: Learning Causal Mechanisms (ICLR invited talk)

Но что такое нейронная сеть? | Глава 1. Глубокое обучение

Но что такое нейронная сеть? | Глава 1. Глубокое обучение