Up to Speed on AI and Deep Learning 2/11/20

AI for AI: Metareasoning for modular computing systems

Microsoft researchers propose a metareasoning approach that can assess a software system pipeline and adjust the parameters of individual modules for proper tradeoff among latency, accuracy, and other factors to ensure optimal operation of the entire system in real time using learning reinforcement. [Read More]


Hey Alexa! Sorry I fooled you …

MIT’s new system TextFooler can trick the types of natural-language-processing systems that Google uses to help power its search results, including audio for Google Home. [Read More]


Intrusion alert: System uses machine learning, curiosity-driven ‘honeypots’ to stop cyberattacks

The Purdue team created a detection system to alert organizations to cyberattacks. The system is called LIDAR—which stands for lifelong, intelligent, diverse, agile and robust. The Purdue system is made up of three main parts: supervised machine learning, unsupervised machine learning and rule-based learning.  [Read More]


Using AI to spot causative relationships in overlapping medical datasets

A combined team of researchers from Babylon Health and University College has created an algorithm that they claim can find causal relationships among information in overlapping medical datasets. They have written a paper describing their algorithm and have uploaded it to the arXiv preprint server. [Read More]


Emerging research

The Costs and Benefits of Goal-Directed Attention in Deep Convolutional Neural Networks. (arXiv:2002.02342v1 [cs.LG])

Motivated by neuroscience research, researchers evaluated a plug-and-play, top-down attention layer that is easily added to existing deep convolutional neural networks (DCNNs). In object recognition tasks, increasing top-down attention has benefits (increasing hit rates) and costs (increasing false alarm rates). [Read More]


Relational Neural Machines. (arXiv:2002.02193v1 [cs.AI])

This paper presents Relational Neural Machines, a novel framework allowing to jointly train the parameters of the learners and of a First–Order Logic based reasoner. Proper algorithmic solutions are devised to make learning and inference tractable in large-scale problems. The experiments show promising results in different relational tasks. [Read More]


Reinforcement Learning-based Fast Charging Control Strategy for Li-ion Batteries. (arXiv:2002.02060v1 [eess.SY])

This paper proposes a fast-charging strategy subject to safety constraints which relies on a model-free reinforcement learning framework. In particular, it focuses on the policy gradient-based actor-critic algorithm, i.e., deep deterministic policy gradient (DDPG), in order to deal with continuous sets of actions and sets. [Read More]


A Neural Approach to Ordinal Regression for the Preventive Assessment of Developmental Dyslexia. (arXiv:2002.02184v1 [cs.LG])

This paper propses a new methodology to assess the risk of DD before students learn to read. For this purpose, a mixed neural model that calculates risk levels of dyslexia from tests that can be completed at the age of 5 years is proposed. This method first trains an auto-encoder, and then combines the trained encoder with an optimized ordinal regression neural network devised to ensure consistency of predictions. [Read More]


AI & ML in Society

The Rise Of The AI Agency In 2020

In a world of increasing information and races to the bottom, AI is more obviously an important competitive advantage for many organizations. [Read More]


Organizations can’t afford to wait for AI talent to come knocking

When we talk about the movement towards artificial intelligence, we tend to assume organizations eventual adoption will be led by a fresh team of newly-appointed experts our co-workers of tomorrow will be Python developers and data scientists. Instead, companies should look at transferable talent under their noses. [Read More]


Google’s New ML Fairness Gym Has A Clear Mission – Track Down Bias & Promote Fairness In AI

Human societies are extremely complex. The cultural, racial and geographical differences around the globe and the lack of curated data make fairness in technology a huge challenge. Now, in an attempt to track the long term societal impacts of artificial intelligence, Google researchers recently released a machine learning fairness gym. [Read More]


AI, 5G, and IoT can help deliver the promise of precision medicine

As AI adoption continues and pairs with faster hardware, more diverse medical devices, and faster connectivity, perhaps we will soon reach a time when no parent ever has to watch an unresponsive child whisked away by ambulance because of adverse reactions that might have been avoided through precision medicine and next-gen technology. [Read More]

Venture Beat

Share this post: