Open in app
Sign inGet started
AAAI 2020 | A Turning Point for Deep Learning?

AAAI 2020 | A Turning Point for Deep Learning?

Geoffrey Hinton, Yann LeCun, and Yoshua Bengio shared a stage in New York on Sunday night at an event organized by AAAI 2020.
Go to the profile of Synced
Synced
Latest
Learning Without Simulations?

Learning Without Simulations?

Using reinforcement learning (RL) to train robots directly in real-world environments has been considered impractical due to the huge…
Go to the profile of Synced
Synced
‘P-Meta’ Learning Approach Boosts Data and Memory Efficiency for On-Device DNN Adaptation in the…

‘P-Meta’ Learning Approach Boosts Data and Memory Efficiency for On-Device DNN Adaptation in the…

Deep learning-powered applications are playing an increasing role in the massive network of interconnected smart devices known as the…
Go to the profile of Synced
Synced
NVIDIA’s Global Context ViT Achieves SOTA Performance on CV Tasks Without Expensive Computation

NVIDIA’s Global Context ViT Achieves SOTA Performance on CV Tasks Without Expensive Computation

Building on the epoch-making performance of transformer architectures in natural language processing (NLP), the vision transformer (ViT)…
Go to the profile of Synced
Synced
CMU’s Novel ‘ReStructured Pre-training’ NLP Approach Scores 40 Points Above Student Average on a…

CMU’s Novel ‘ReStructured Pre-training’ NLP Approach Scores 40 Points Above Student Average on a…

The paradigms for natural language processing (NLP) are rapidly evolving — from fully supervised learning to pretraining and fine-tuning…
Go to the profile of Synced
Synced
Google’s Novel Lossy Compression Method Targets Perfect Realism with Only a Single Diffusion Model

Google’s Novel Lossy Compression Method Targets Perfect Realism with Only a Single Diffusion Model

Lossy compression is a data compression approach with a trade-off: it enables high compression but incurs some loss of information, i.e…
Go to the profile of Synced
Synced
Allen AI & UW Propose Unified-IO: A High-Performance, Task-Agnostic Model for CV, NLP, and…

Allen AI & UW Propose Unified-IO: A High-Performance, Task-Agnostic Model for CV, NLP, and…

Building a general-purpose unified model that can solve diverse tasks in different modalities while maintaining high performance is a…
Go to the profile of Synced
Synced
Intelligent Mutations in Genetic Programming: OpenAI Proposes Evolution Through Large Models

Intelligent Mutations in Genetic Programming: OpenAI Proposes Evolution Through Large Models

Large-scale language models (LLMs) have achieved impressive performance in automated code generation by bootstrapping human knowledge and…
Go to the profile of Synced
Synced
A WaveNet Rival? Stanford U Study Models Raw Audio Waveforms Over Contexts of 500k Samples

A WaveNet Rival? Stanford U Study Models Raw Audio Waveforms Over Contexts of 500k Samples

The effective modelling of long-term dependencies enables conditioning new model outputs on previous inputs and is critical when dealing…
Go to the profile of Synced
Synced
DeepMind Boosts RL Agents’ Retrieval Capability to Tens of Millions of Pieces of Information

DeepMind Boosts RL Agents’ Retrieval Capability to Tens of Millions of Pieces of Information

The conventional approach for improving the decision-making of deep reinforcement learning (RL) agents is to gradually amortize the useful…
Go to the profile of Synced
Synced
About SyncedReviewLatest StoriesArchiveAbout MediumTermsPrivacy