Description and Goals:

The world currently faces a number of pressing global challenges. Energy, Healthcare, Sustainability, Environment, and Education are exemplary areas where cross-disciplinary research is assembling large teams of people from around the world to address urgent global concerns. We believe that machine learning has much to offer as a key component of these cross-disciplinary efforts.  The goal of this workshop is to bring together researchers working in different technical areas of machine learning to facilitate interactions among them, as well as to connect researchers with experts in these domains.

While the machine learning community agrees that our field is highly relevant, even critical, in solving these global challenges, most of the current research focus in machine learning does not directly aim to solve these issues.  We posit that this gap exists for the following three reasons:
1) Lack of awareness of these problems at a level of technical detail that allows them to be formulated crisply in machine learning terms.
2) Limited access to domain experts who can provide a practical perspective on existing problems and provide a new set of research directions to the community.
3) Recent surge of interest in applied machine learning on problems related to Web Search, Advertising and Recommender Systems arguably due to easier availability of data and business interests around monetizing the web.
The goals of this workshop are to:

1) Bring together researchers from different areas of machine learning working on topics that are technically unrelated but can work together to also solve one or more of the global challenges.

2) Attract experts working in the areas of these global challenges from both academia and industry who have the domain expertise and access to useful sources of data and problems but not necessarily the expertise in ML to solve them effectively. This group would otherwise not attend ICML and we believe that it is essential for ML research community to interact with them.

Workshop Topics and Structure

We are focusing this workshop on three global challenges: Sustainability, Healthcare, and Education & Infrastructure for the developing world. We plan on having three sessions. Each session will include invited talks from domain experts in each field, panel of machine learning and domain experts, and a poster session. 
  1. Environment

    Many of the most pressing global challenges that will face society in the next century involve the Environment, and Climate Change, and approaches to these problems often entail work towards Sustainability.  Half of the workshop will be devoted to addressing pressing Environmental issues, to which Machine Learning is poised to make impactful contributions.  We will accept submissions in related areas, including but not limited to:

    • Environment 
    • Sustainability
    • Climate Science
    • Climate Change Impacts
    • Energy

  2. Healthcare

    Healthcare systems around the world are struggling to keep up with patient needs, and improve quality of care while reducing costs at the same time. At the same time, more and more data is being captured around healthcare processes in the form of Electronic Medical Records (EMR), health insurance claims, medical imaging databases, disease registries, spontaneous reporting sites, and clinical trials. As this data gets collected, government regulations are requiring healthcare providers to not only store it in an electronic format but also use it in meaningful ways. Using this data in an effective way to improve quality of care and reduce costs requires the use of novel machine learning algorithms. We will accept submissions in related areas, including but not limited to:
    • Meaningful use of healthcare data for improved patient care 
    • Pattern detection and hypothesis generation from observational data
    • Evolutionary and longitudinal patient and disease models
    • Mining knowledge from medical imaging data
    • Medical fraud detection
    • Bio-surveillance
    • Post-market surveillance of medical interventions
    • Text mining - mining free text in electronic medical records
    • Improving Clinical trial management and design

  3. Education, Infrastructure & other challenges for the developing world       
        Can machine learning and data mining help improve educational systems and the process of human learning? Can we build low-cost, efficient platforms for improving infrastructure in the developing world using insights drawn from data?

Impact and expected outcomes:

We believe that this workshop can have a high impact both within the machine learning community as well as in the larger world. The issues we propose to tackle in this workshop are major challenges that the world is facing today, and focusing the attention of the machine learning community on these problems is critical for our future. In addition to having a direct impact in terms of solving important problems, this workshop is also targeted at making researchers and grad students aware of applied machine learning in areas other than search, advertising, and e-commerce which have been dominating the machine learning community in the recent past. Finally, we believe that bringing together the machine learning community to discuss and solve these global challenges increases the visibility of machine learning in the mainstream media and public perception which is beneficial for our field and community.

Related Events

NIPS 2010 Workshop on Predictive Models in Personalized Medicine
ICDM 2010 Workshop on Knowledge Discovery from Climate Data
NIPS 2009 Mini-symposium on Machine Learning for Sustainability
ICML 2008 Workshop on Machine Learning in Healthcare Applications