SMiLe CLiNiC research aims to develop flexible yet efficient probabilistic learning methods that can take into account diverse statistical features of real-world data. While machine learning methods have emerged as one of the most promising statistical frameworks for addressing real-world challenges, many of the current models and algorithms are too restrictive for capturing complex tasks, and are challenging for massive scale applications. In the latest study by IDC (J. Gantz and D. Reinsel. The Digital Universe in 2020: Big Data, Bigger Digital Shadows, and Biggest Growth in the Far East.), it is forecasted that 40 ZettaBytes of digital content were created or replicated by 2020, and as much as 33% of the digital universe will contain information that might be valuable if analysed.

Complex statistical features are commonplace for those data; SMiLe CLiNiC focuses on the heterogeneity or multi-modality of the data. Phrased differently data that has textual, imaging, and network relationship representations and comes from multiple sources. Being in a real-world setting, the multi-modality challenge turns into a multi-faceted problem containing the following three major sub-problems: i) dynamic time evolution (data is collected over a long time period); ii) rich interdependency structures (in text, not only words but also how words combine into a sentence carry relevant information); iii) output inconsistencies (information mismatch between or within data sources). Scientific and societal challenges of the 21st century are to draw insights and make predictions to support knowledge creation from this massive amount of data.

SMiLe CLiNiC is part of the Sussex Informatics department, affiliated with Data Science @ Sussex and based in the Sussex campus set on the edge of the beautiful South Downs National Park. Key facts about Sussex: 1) Top 10 in the UK - 60th in the world for research influence (Times Higher Education, World University Rankings 2013-2014), 2) The sunniest part of UK, 3) Less than half an hour by cycle to the Brighton Beach, 4) The 6th safest university cities and towns.

• We are now Predictive Analytics (#WeArePAL) Laboratory https://wearepal.ai/
• Sep 2017: A paper on fair machine learning models is accepted at NIPS 2017.
• Jul 2017: Announcing Post-Doc and PhD positions in ethical machine learning.
• May 2017: Awarded an EPSRC grant on EthicalML: Injecting Ethical and Legal Constraints into Machine Learning Models.
• From 20 May 2017 until 10 June 2017, Novi is in Moscow for International Lab of Deep Learning and Bayesian Methods.
• Novi is an Area Chair for NIPS 2017.
• One paper is accepted at AISTATS 2017 (Congratulations to Pietro!) and another one at ICML 2017.