Artificial Intelligence is a very wide term, so we’ve picked some interesting research papers from 2018, plus the highlights of 2017, which cover a diverse range of themes and topics around the core idea of Artificial Intelligence.
Deep machine learning provides state-of-the-art performance in image-based plant phenotyping (link)
By Michael P. Pound, Jonathan A. Atkinson, Alexandra J. Townsend, Michael H. Wilson, Marcus Griffiths, Aaron S. Jackson, Adrian Bulat, Georgios Tzimiropoulos, Darren M. Wells, Erik H. Murchie, Tony P. Pridmore, Andrew P. French
‘In plant phenotyping, it has become important to be able to measure many features on large image sets in order to aid genetic discovery. The size of the datasets, now often captured robotically, often precludes manual inspection, hence the motivation for finding a fully automated approach. Deep learning is an emerging field that promises unparalleled results on many data analysis problems. Building on artificial neural networks, deep approaches have many more hidden layers in the network, and hence have greater discriminative and predictive power…’
Political Economy as Both a Challenge to, and a Source of, Human Engagement in the Future Economy (link) .
‘Discussion of: Artificial Intelligence and the Future of Growth, by Philippe Aghion, Benjamin Jones and Charles E. Jones. By Patrick Francois Vancouver School of Economics, University of British Columbia Canadian Institute For Advanced Research’
How Artificial Intelligence and Machine Learning Can Impact Market Design (link)
‘In complex environments, it is challenging to learn enough about the underlying characteristics of transactions so as to design the best institutions to efficiently generate gains from trade. In recent years, Artificial Intelligence has emerged as an important tool that allows market designers to uncover important market fundamentals, and to better predict fluctuations that can cause friction in markets. This paper offers some recent examples of how Artificial Intelligence helps market designers improve the operations of markets, and outlines directions in which it will continue to shape and influence market design.’
Understanding the nature of the human mind via simplification and integration across artificial intelligence and related fields (link)
By J Gerard Wolff
‘Towards an understanding of the nature of the human mind, this article summarises the aims and results of an extended programme of research developing the SP theory of intelligence and its realisation in the SP computer model. The overall aim of the research is simplification and integration of observations and concepts across artificial intelligence, mainstream computing, mathematics, and human learning, perception, and cognition. Perhaps the most significant outcome of this research is the discovery that a concept of SP-multiplealignment, borrowed and adapted from the concept of ‘multiple sequence alignment’ in bioinformatics, has proved to be a key to the modelling of diverse aspects of human intelligence and the representation of diverse kinds of knowledge.’
Waiting for a sales renaissance in the fourth industrial revolution: Machine learning and artificial intelligence in sales research and practice (link)
‘Experts have suggested that the next few decades will herald the fourth industrial revolution. The fourth industrial revolution will be powered by digitization, information and communications technology, machine learning, robotics and artificial intelligence; and will shift more decision-making from humans to machines. The ensuing societal changes will have a profound impact on both personal selling and sales management research and practices. In this article, we focus on machine learning and artificial intelligence (AI) and their impact on personal selling and sales management. We examine that impact on a small area of sales practice and research based on the seven steps of the selling process. Implications for theory and practice are derived.’
Artificial Intelligence and Foreign Policy
‘The plot-lines of the development of Artificial Intelligence (AI) are debated and contested. But it is safe to predict that it will become one of the central technologies of the 21st century. It is fashionable these days to speak about data as the new oil. But if we want to “refine” the vast quantities of data we are collecting today and make sense of it, we will need potent AI. The consequences of the AI revolution could not be more far reaching. Value chains will be turned upside down, labor markets will get disrupted and economic power will shift to those who control this new technology. And as AI is deeply embedded in the connectivity of the Internet, the challenge of AI is global in nature. Therefore it is striking that AI is almost absent from the foreign policy agenda.’
By Jun-Yan Zhu, Taesung Park, Phillip Isola, Alexei A. Efros (from Berkeley AI Research)
Goal: Learn to translate between unpaired sets of images.
‘The authors begin with two sets of images from different domains, e.g. of horses and zebras, and learn two translation networks: one that translates horse images to zebra images and another that translates zebra images to horse images. Each translator performs a sort of style transfer, but rather than targeting the style of a single image, the network discovers the aggregate style of a set of images.’
By Ashish Shrivastava, Tomas Pfister, Oncel Tuzel, Josh Susskind, Wenda Wang, Russ Webb (of Apple)
Goal: Use real-world images to make simulated training data more useful for real-world applications.
‘Collecting real-world data can be difficult and time-consuming. As such, many researchers will frequently use simulation toolsTools like the OpenAI gym are particularly useful for training data-hungry deep reinforcement learning agents., which are capable of generating nearly infinite amounts of labeled training data. However, most simulated data is not sufficiently realistic for training deep learning systems that operate on real-world data.’
The best Deep Learning papers of 2017
- In this ‘2017 year in review’ article you will find some great papers to read.
- These were the most read Deep Learning papers of 2017.
Video of the year?