Top news of the week: 26.03.2020.

#MSRPodcast #CHI2020 #AI #machinelearning #research

Research

On Mar 23, 2020
@mxlearn shared
[P] Visualizing and explaining Deep Learning Antibiotics https://t.co/rJgX1s7B0n
Open

[P] Visualizing and explaining Deep Learning Antibiotics

[P] Visualizing and explaining Deep Learning Antibiotics

A few weeks ago we saw a paper detailing the discovery of antibiotic properties of *Halicin* by means of a deep neural network. I read both that...

On Mar 25, 2020
@hardmaru shared
RT @jeffclune: Secretly our Enhanced POET project was an attempt to move closer to making this picture from @hardmaru's ES blog post a reality. We use the same algorithm, and now have the cliff jumping! Next up: automating the generation of environments with parachutes! https://t.co/07PwX4i0ym https://t.co/3YCA8Py9og
Open
On Mar 21, 2020
@kchonyc shared
@jeffclune ha doesn't it remind me of your paper :) https://t.co/hgbTmbiXkw
Open

The evolutionary origins of modularity

The evolutionary origins of modularity

Results After 25 000 generations in an unchanging environment (L-AND-R), treatments selected to maximize performance and minimize connection costs (P&CC) produce significantly …

On Mar 24, 2020
@tqchenml shared
RT @octoml: Check out our first blog post describing the "Octomizer" we are building to make it easier (and faster) to deploy ML apps. Also read our take on how ML/DL software foundations are broken and what we are doing to fix it. https://t.co/7yu72iG1le
Open

OctoML ≔ Easier machine learning

OctoML ≔ Easier machine learning

Machine learning and deep learning (ML/DL) are making large impacts across the computing field and the horizon is bright with the rapidly…

On Mar 25, 2020
@MSFTResearch shared
Rangan Majumder explores how fusing AI’s deep learning and symbolist camps will help make the world smarter and more productive. @RangThang also shares trends changing his team's approach to AI and their deep learning moon shot, Project Turing: https://t.co/x4plGtSHnc #MSRPodcast
Open

Microsoft’s AI Transformation, Project Turing and smarter search with Rangan Majumder

Microsoft’s AI Transformation, Project Turing and smarter search with Rangan Majumder

Episode 112 | March 25, 2020 - Rangan Majumder is the Partner Group Program Manager of Microsoft’s Search and AI, and he has a simple goal: to make the world smarter and more productive. ...

On Mar 23, 2020
@agibsonccc shared
RT @PyTorch: Captum is a library for model interpretability. Its algorithms include integrated gradients, conductance, SmoothGrad and VarGrad, and DeepLift. Learn more: https://t.co/IVdye9smGi
Open

Introduction to Captum — A model interpretability library for PyTorch

Introduction to Captum — A model interpretability library for PyTorch

A model interpretability library for PyTorch that allow us to understand the importance of input features, and hidden neurons and layers.

On Mar 24, 2020
@MSFTResearch shared
Internet-delivered Cognitive Behavioral Therapy is a means for people with mental illness to get needed help alongside trained coaches. Learn how researchers are enlisting machine learning to find the most successful techniques these coaches use: https://t.co/HDhbIgfiYD #CHI2020
Open

Data-driven insights for more effective, personalized care in online mental health interventions

Data-driven insights for more effective, personalized care in online mental health interventions

Internet-delivered Cognitive Behavioral Therapy is a means for people with mental illness to get needed help alongside trained coaches. Discover how researchers are enlisting machine ...

On Mar 25, 2020
@hardmaru shared
RT @Synced_Global: Researchers from Google Brain Tokyo and Google Japan have proposed a novel approach that helps guide reinforcement learning (RL) agents to what’s important in vision-based tasks. Read the full story at https://t.co/x5oXm0B5Dj #machinelearning #AI #research https://t.co/nakmqlhXxh
Open

Google Introduces Neuroevolution for Self-Interpretable Agents

Google Introduces Neuroevolution for Self-Interpretable Agents

Researchers have proposed a novel approach that helps guide reinforcement learning (RL) agents to what’s important in vision-based tasks.