Amazon, Facebook, IBM, Microsoft and Google's DeepMind subsidiary have agreed to create a non-profit organisation that will work to advance public understanding of artificial intelligence technologies (AI) and come up with best practices for the sector. The alliance, dubbed the Partnership on Artificial Intelligence to Benefit People and Society, has a stated aim to "conduct research, recommend best practices and publish research under an open licence in areas such as ethics, fairness and inclusivity; transparency, privacy, and interoperability; collaboration between people and AI systems;

Officially titled the "Partnership on Artificial Intelligence to Benefit People and Society," the group's stated goals are to pool resources and develop interoperability for the future of AI technology. At this time, the group has declared that it does not intend to become a governmental lobbyist group. To meet its goals, the organization anticipates it will "host discussions, commission studies, write and distribute reports on critical topics, and seek to develop and share best practices and standards for industry." Additionally, the group states that it will "conduct outreach with the public and across the industry on topics related to advancing better understanding of AI systems and the potential applications and implications of this technology as they arise." The founding corporate members of the group are Amazon, DeepMind, Facebook, Google, IBM, and Microsoft with each company holding one spot on the board of directors.

Media attending the event included BBC News, The Guardian, The Wall Street Journal, Bloomberg, VentureBeat, Digital Trends, Financial Times, Ars Technica and more. News coverage focused on a range of topics, exploring advancements in robotics, chatbot personalities, machine vision for understanding differences in language and culture, as well as startup acquisitions and funding. We've shared just a few of the great articles from the summit below. Why Data is the New Coal - The Guardian Deep learning needs to become more efficient if it is going to move from using data to categorise images of cats to diagnosing rare illnesses. Alex Hern reports on revelations in this area from speaker Neil Lawrence, the newly appointed Senior Principal Scientist at Amazon.

In this paper, we show an approach to build deep learning algorithms for recognizing signals in distributed fiber optic monitoring and security systems for long perimeters. Synthesizing such detection algorithms poses a nontrivial research and development challenge, because these systems face stringent error (type I and II) requirements and operate in difficult signal-jamming environments, with intensive signal-like jamming and a variety of changing possible signal portraits of possible recognized events. To address these issues, we have developed a two-level event detection architecture, where the primary classifier is based on an ensemble of deep convolutional networks, can recognize 7 classes of signals and receives time-space data frames as input. Using real-life data, we have shown that the applied methods result in efficient and robust multiclass detection algorithms that have a high degree of adaptability.

Convolutional networks have marked their place over the last few years as the best performing model for various visual tasks. They are, however, most suited for supervised learning from large amounts of labeled data. Previous attempts have been made to use unlabeled data to improve model performance by applying unsupervised techniques. These attempts require different architectures and training methods. In this work we present a novel approach for unsupervised training of Convolutional networks that is based on contrasting between spatial regions within images. This criterion can be employed within conventional neural networks and trained using standard techniques such as SGD and back-propagation, thus complementing supervised methods.

Machine learning (ML) models may be deemed confidential due to their sensitive training data, commercial value, or use in security applications. Increasingly often, confidential ML models are being deployed with publicly accessible query interfaces. ML-as-a-service ("predictive analytics") systems are an example: Some allow users to train models on potentially sensitive data and charge others for access on a pay-per-query basis. The tension between model confidentiality and public access motivates our investigation of model extraction attacks. In such attacks, an adversary with black-box access, but no prior knowledge of an ML model's parameters or training data, aims to duplicate the functionality of (i.e., "steal") the model. Unlike in classical learning theory settings, ML-as-a-service offerings may accept partial feature vectors as inputs and include confidence values with predictions. Given these practices, we show simple, efficient attacks that extract target ML models with near-perfect fidelity for popular model classes including logistic regression, neural networks, and decision trees. We demonstrate these attacks against the online services of BigML and Amazon Machine Learning. We further show that the natural countermeasure of omitting confidence values from model outputs still admits potentially harmful model extraction attacks. Our results highlight the need for careful ML model deployment and new model extraction countermeasures.

The major roadblock is applying the right set of tools, which can pull powerful insights from this stockpile of data. But first, a big data system requires identifying and storing of digital information (lots of!!). Using Machine learning and Artificial Intelligence algorithms, businesses can optimize and uncover new statistical patterns which form the backbone of predictive analytics. Organization with huge data can begin analytics. Before beginning data scientists should make sure that predictive analytics fulfills their business goals and is appropriate for the big data environment.

I am guessing that the reason Apple and Musk are not involved is probably because they both have their own projects on the go. Or at least for sure Elon does, a consortium called OpenAI openai.com/blog/ . They are not afraid of going it alone, and they have the funds and other resources to do so. If its at all possible… and i think it is myself.. these are probably the ones who will find a way.. Which brings the next question, and most important .. SHOULD we be doing this sort of work? Thats one that should be asked before we go too far ahead a lot of people think.

Claire Bretton is one of the co-founders of daco.io, a startup that is developing a unique tool to track competition thanks to deep learning. Earlier, she was a manager in a top strategy consulting firm based in Paris. She holds a master's degree from ESCP Europe. One of the biggest challenges of the 21st century is to make computers more similar to the human brain. We want them to speak, understand and solve problems -- and now we want them to see and recognize images.

For the 22nd year, Gartner has released its much-discussed hype cycle report on emerging technologies, "providing a cross-industry perspective on the technologies and trends that business strategists, chief innovation officers, R&D leaders, entrepreneurs, global market developers and emerging-technology teams should consider in developing emerging-technology portfolios." Reacting to last year's hype cycle report (see below), I made the following comment: Machine learning is making its first appearance on the chart this year, but already past the peak of inflated expectations. A glaring omission here is "deep learning," the new label for and the new generation of machine learning, and one of the most hyped emerging technologies of the past couple of years. This year, Gartner has moved machine learning back a few notches, putting it at the peak of inflated expectations, still with 2 to 5 years until mainstream adoption. Is machine learning an emerging technology and is there a better term to describe what most of the hype is about nowadays in tech circles?