- Deepmind: Agents that imagine a plan https://deepmind.com/blog/agents-imagine-and-plan/
- Densely connected CNNs : https://arxiv.org/pdf/1608.06993.pdf
- Qualcomm releases its Neural Processing Engine SDK to developers via their developer network
- Caffe2 adds RNN support
- Tenssorfire by webGL: https://tenso.rs/demos/fast-neural-style
- Deepmind: Producing flexible behaviours in simulated environments https://deepmind.com/blog/producing-flexible-behaviours-simulated-environments/
- Deepmind: Distral: Robust Multitask Reinforcement Learning https://arxiv.org/pdf/1707.04175.pdf
- OpenAI: New class of reinforcement learning algorithms, Proximal Policy Optimization (PPO) https://blog.openai.com/openai-baselines-ppo/
- OpenAI: Creating Robust adverrsarial examples https://blog.openai.com/robust-adversarial-inputs/
- OpenAI Baselines: high-quality implementations of reinforcement learning algorithms
- Exploring large datasets with PAIR
- CVPR kickedoff at HonoLulu
- Click Professional photos with ML https://google.github.io/creatism/
- Gradient Ventures by Google
- PixelGAN autoencoder: https://arxiv.org/abs/1706.00531
- Drive.ai raises 50M$ and adds Andrew Ng
- Fake AV Generation with GANs: https://www.economist.com/news/science-and-technology/21724370-generating-convincing-audio-and-video-fake-events-fake-news-you-aint-seen
- Nvidia partners with volvo to develop self driving cars by 2021
- Sony opensources Nnabla: Neural network libraries
- Open AI open sources robotic simulation library based on MoJuCo engine : https://blog.openai.com/faster-robot-simulation-in-python
- Performance RNN by Magenta: LSTM-based recurrent neural network designed to model polyphonic music with expressive timing and dynamics
- New BAIR Post on neural module networks(NMNs).
- MultiModel: Multi-Task Machine Learning Across Domains https://research.googleblog.com/2017/06/multimodel-multi-task-machine-learning.html
- Deep reinforcement learning from human preferences https://arxiv.org/abs/1706.03741
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour : Caffe2-based system trains ResNet-50 with a minibatch size of 8192 on 256 GPUs in one hour,
- Karpathy joins Tesla
- Andrew NG Launches deeplearning.ai
- Apple is working on software for self-driving cars
- Apple releases CoreML
- TensorFlow object detection API https://research.googleblog.com/2017/06/supercharge-your-computer-vision-models.html
- T2T: Tensor2Tensor Transformers
-
End-to-end optimization of goal-driven and visually grounded dialogue systems https://arxiv.org/abs/1703.05423
-
Learning to Skim Text https://arxiv.org/abs/1704.06877
-
Google updates Maps with deep learning and open sources the code of the model https://github.com/tensorflow/models/tree/master/attention_ocr https://research.googleblog.com/2017/05/updating-google-maps-with-deep-learning.html
-
AI is being used by GE to optimize windfarms in Japan http://asia.nikkei.com/Business/Companies/AI-to-propel-wind-farm-efficiency-in-Japan
- ICLR happened!!! and videos are available at https://www.facebook.com/pg/iclr.cc/videos/
- FAIR released ParlAI( https://parl.ai)
- TensorFlow v1.1.0 released
- FAIR fasttext library now available on smaller devices https://code.facebook.com/posts/162370720961195/
- IPython 6.0 is out- without python 2 support!
- Jupyter 5.0 released
- ipyvolume 0.3 released
- Elon Musk launches a Brain computer interface startup called neuralink
- Launched by facebook on F8
- https://Caffe2.ai
- java API ssupport for windows, D, 2D and 3D Fourier transform ops for real signals to tf.spectral, Keras 2 API to contrib
- https://github.com/tensorflow/tensorflow/releases
- what is sonnet? Library which enables deep mind's best practices internally, easy switch to different models while conductiong experiments
- https://deepmind.com/blog/open-sourcing-sonnet/
- https://github.com/deepmind/sonnet
- https://github.com/deepmind/dnc
- 91.8% accuracy versus the previous best of 90.2% on the Stanford Sentiment Treebank
- https://blog.openai.com/unsupervised-sentiment-neuron/
- https://github.com/openai/generating-reviews-discovering-sentiment
- Neural Audio Synthesis library developed by Boogle brain team and Deepmind
- https://magenta.tensorflow.org/nsynth
- Tensorflow integration with tf.keras
- The legacy layers MaxoutDense, TimeDistributedDense, and Highway have been permanently removed.
- https://blog.keras.io/introducing-keras-2.html
- Google reserach journal
- Features: more ofcus on understanding, Visualization of algos and Code
- More focus on transperency
- First edition features: Handwriting with ANN, Attention and Augmented RNNs, t-Distributed Stochastic Neighbor Embedding (t-SNE) plots
- http://distill.pub/
- guessing geolocation with photos
- Text-to-speech system constructed entirely from deep neural networks as compared to modern TTS which arecomplex, multi-stage processing pipelines, each of which may rely on hand-engineered features and heuristics
- Less labour intensive and easier to program
- http://research.baidu.com/deep-voice-production-quality-text-speech-system-constructed-entirely-deep-neural-networks/