We Research and Build
TurinTech is built upon years of research in the space of AI, combining the experience and expertise of the best data scientists, mathematicians and software engineers. Our world-class research has been published in some of the top journals and presented at world leading conferences such as ICSE Conference.

Our research approach
We use our knowledge to solve the problems we see in the world.
Our research philosophy is quite straightforward; we want to use our knowledge and expertise to solve the problems we see in the world. TurinTech was formed to find answers to a problem our co-founders encountered in their everyday work, and our research approach has not changed much since conceptualisation. We want to provide sustainable solutions to our users’ biggest pain points. With this in mind, we engage with our users regularly, discuss concerns, and incorporate their feedback into our platform, so that our research truly solves their problems.

As we make improvements to our platform, we also make it a point to independently research the machine learning space. Guided by intellectual curiosity, we aim to understand artificial intelligence better so that our users, researchers, and other stakeholders can benefit from being at the forefront of innovation. Our innovation-led approach allows customers to improve their present workflows, while also preparing for the future. We make sure that our clients are aware of the trends, opportunities, as well as challenges to expect in the years to come, so the future never takes them by surprise.
Our researchers have gone through rigorous academic training, and most of us have also worked across various organisations in the financial sector. We have studied the theoretical underpinnings of machine learning and AI, while we also understand the practical realities of building AI solutions in commercial development environments. Combined together, this expertise place us in a unique position to develop robust technologies that suit fast-paced industry environments.

Research Papers

RESEARCH
We introduce ARTEMIS, a multi-objective, cloud-based search-based optimisation framework that automatically finds optimal, tuned Darwinian Data Structure, then automatically changes an application to use that DDS.
Michail Basios , Lingbo Li , Fan Wu , Leslie Kanthan , Earl T. Barr, 2018
RESEARCH
In this paper, we propose a new tree-model explanation approach for model selection.
Fan Fang, Carmine Ventre, Lingbo Li, Leslie Kanthan, Fan Wu, Michail Basios, 2020
RESEARCH
Our proposed code optimisation solution, called Artemis++, tries to optimise inefficient data structures or library interfaces with automatic exploration and transformation of data structures to optimise software performance.
Wu, F, 2017
RESEARCH
We conducted large scale experimental research on the performance of two state-of-the-art code clone detection techniques, SourcererCC and AutoenCODE, on both open source projects and an industrial project written in the Scala language.
Wahidur Rahman, Yisen Xu, Fan Pu; Jifeng Xuan; Xiangyang Jia, Michail Basios, Leslie Kanthan, Lingbo Li, Fan Wu, Baowen Xu, 2020

RESEARCH
This is one of the most influencial and first surveys in the area of cryptocurrency trading which explains in details how machine learning techniques can be used for algo trading.
Fan Fang, Carmine Ventre, Michail Basios, Leslie Kanthan, Lingbo Li, David Martinez-Regoband, Fan Wu, 2020
RESEARCH
We introduce Memory Mutation Testing, proposing 9 Memory Mutation Operators each of which targets common forms of memory fault. We compare Memory Mutation Operators with traditional Mutation Operators, while handling equivalent and duplicate mutants.
Fan Wu, Jay Nanavati, Mark Harman Yue Jia, Jens Krinke, 2017

SHORT PAPER
“This paper presents a brief outline of a higher-order mutation based framework for Genetic Improvement (GI). We argue that search-based higher-order mutation testing can be used to implement a form of genetic programming (GP) to increase the search granularity and testability of GI.”
Yue Jia, Fan Wu, Mark Harman, Jens Krinke 2015
RESEARCH
We present a search based testing system that automatically explores the space of all possible GUI event interleavings. Search guides our system to novel crashing sequences using Levenshtein distance and minimises the resulting fault-revealing UI sequences in a post-processing hill climb.
Haitao Dan, Mark Harman, Jens Krinke, Lingbo Li, Alexandru Marginean, Fan Wu, 2014
RESEARCH
A new more efficient genetic based multiobjective algorithm is proposed to optimise software requirements for next release.
Lingbo Li, Mark Harman, Emmanuel Letier, Yuanyuan Zhang, 2014

RESEARCH
We introduce a mutation-based approach to automatically discover and expose ‘deep’ (previously unavailable) parameters that affect a program’s runtime costs. These discovered parameters, together with existing (‘shallow’) parameters, form a search space that we tune using search-based optimisation in a bi-objective formulation that optimises both time and memory consumption.
Fan Wu, Westley Weimer, Mark Harman, Yue Jia, Jens Krinke, 2015
RESEARCH
In this paper is was shown how deep learning approaches can be used to predict the direction of the mid-price changes of crypto assets.
Fan Fang, Waichung Chung, Carmine Ventre, Michail Basios, Leslie Kanthan, Lingbo Li, Fan Wu, 2020

RESEARCH
In this paper, we discuss an automated approach for exploring API equivalence and a framework to synthesise semantically equivalent programs.
Fan Wu, Westley Weimer, Mark Harman, Yue Jia, Jens Krinke, 2015

RESEARCH
The thesis applies Mutation Operators to automatically modify the source code of the target software. After a prior sensitivity analysis on First Order Mutants, “deep” (previously unavailable) parameters are exposed from the most sensitive locations, followed by a bi-objective optimisation process to fine tune them together with existing (“shallow”) parameters. The objective is to improve both time and memory resources required by the computation.
Wu, F, 2017

RESEARCH
We introduce mutation-aware fault prediction, which leverages additional guidance from metrics constructed in terms of mutants and the test cases that cover and detect them.
David Bowes, Tracy Hall, Mark Harman, Yue Jia, Federica Sarro, Fan Wu, 2016

ARTICLE
This paper presents a brief outline of an approach to online genetic improvement. We illustrate our proposed approach with a ‘dreaming smart device’ example that combines online and offline machine learning and optimisation.
Mark Harman, Yue Jia, William B. Langdon, Justyna Petke, Iman Hemati Moghadam, Shin Yoo, Fan Wu, 2014
CONFERENCE PAPER
A framework that simulates markets and the behaviour of its actors using decision making AI agents.
uhong Liu, Maria Polukarov, Carmine Ventre, Lingbo Li, Leslie Kanthan, 2021

RESEARCH
In this paper, we introduce an intelligent evolutionary optimisation algorithm which applies machine learning technique to the traditional evolutionary algorithm to accelerate the overall optimisation process of tuning machine learning models in classification problems.
Yuxi Huan, Fan Wu, Michail Basios, Leslie Kanthan, Lingbo Li, Baowen Xu, 2020
RESEARCH
In this work, we analyze and present the characteristics of the cryptocurrency market in a high-frequency setting.
Fan Fang, Waichung Chung, Carmine Ventre, Michail Basios, Leslie Kanthan, Lingbo Li & Fan Wud Turing, 2021
RESEARCH
Our novel code optimisation approach applied to optimise the performance of the popular Google Guava Library. Winner of the seach based software engineering challenge award.
Michail Basios, Lingbo Li, Fan Wu, Leslie Kanthan, Earl T. Barr, 2017
RESEARCH
In this paper, we investigate the use of higher-order functions in Scala programs.
Yisen Xu, Fan Wu, Xiangyang Jia, Lingbo Li & Jifeng Xuan, 2020
RESEARCH
This paper introduces a Higher Order Mutation based approach for Genetic Improvement of software, in which the code modification granularity is finer than in previous work while scalability remains. The approach applies the NSGAII algorithm to search for higher order mutants that improve the non-functional properties of a program while passing all its regression tests.
Fan Wu, Mark Harman, Yue Jia, Jens Krinke, 2016

RESEARCH
This paper introduces 9 Memory Mutation Operators targeting common memory faults and two new testing criteria, the Memory Fault Detection and the Control Flow Deviation criteria to augment the traditional strong mutation testing criterion.
Jay Nanavati, Fan Wu, Mark Harman, Yue Jia, Jens Krinke, 2015
RESEARCH
Determining which functional components should be integrated to a large system is a challenging task, when hardware constraints, such as available memory, are taken into account. We formulate such problem as a multi-objective component selection problem, which searches for feature subsets that balance the provision of maximal functionality at minimal memory resource cost.
Lingbo Li, Mark Harman, Fan Wu, Yuanyuan Zhang, 2015

RESEARCH
A decision support framework (METRO) was proposed that handles the Next Release Problem (NRP) by managing better algorithmic and requirements uncertainty.
Lingbo Li; Mark Harman; Fan Wu; Yuanyuan Zhang, 2012
CONFERENCE PAPER
A new decision support framework for analysing uncertainty in requirements selection and optimisation problem.
Lingbo Li, 2016