View profile

Weekly Machine Learning #44



November 3 · Issue #44 · View online
All change is not growth, as all movement is not forward. Ellen Glasgow

Articles:mixupは簡単ですが有効な正則化方法で、試してみたブログも続々と出てきています。Understanding Hidden Memories~は、CNNに比べて学習内容が把握しにくいRNNにとって貴重なアイデアです。TreeQN はシミュレーション機構をネットワークに組み込むといった趣で、自己モデルを利用したフィードバックを得るという点ではAlphaGo Zeroに近いと思いました。

Google Developers Blog: Eager Execution: An imperative, define-by-run interface to TensorFlow
The State of ML and Data Science 2017 | Kaggle
PaintsChainer -線画自動着色サービス-
Release Keras 2.0.9 · fchollet/keras · GitHub
[1710.09412v1] mixup: Beyond Empirical Risk Minimization
[1710.10777] Understanding Hidden Memories of Recurrent Neural Networks
[1710.10280] One-shot and few-shot learning of word embeddings
[1710.10710] On Pre-Trained Image Features and Synthetic Images for Deep Learning
[1710.11417] TreeQN and ATreeC: Differentiable Tree Planning for Deep Reinforcement Learning
[1711.00155] Neural Wikipedian: Generating Textual Summaries from Knowledge Base Triples
A Visual Guide to Evolution Strategies | 大トロ
How to colorize black & white photos with just 100 lines of neural network code
Deep Learning以外の手法でユニクロコンに参戦してみた (Kaggle Meetup #3) // Speaker Deck
GitHub - jivoi/awesome-ml-for-cybersecurity: Machine Learning for Cyber Security
Extracting Tasks from Emails: first challenges – Rodrigo Alarcón – Medium
QAngaroo Reading Comprehension with Multiple Hops
GitHub - Franck-Dernoncourt/pubmed-rct: PubMed 200k RCT dataset: a large dataset for sequential sentence classification.
ポーカーAIの最新動向 20171031
GitHub - google/bottery
Did you enjoy this issue?
Become a member for $5 per month
Don’t miss out on the other issues by piqcy
You can manage your subscription here
If you were forwarded this newsletter and you like it, you can subscribe here.
Powered by Revue