Skip to main content
School of Electronic Engineering and Computer Science

EECS Research Week 2020

The School of Electronic Engineering and Computer Science at Queen Mary University of London delivers world-class electronic engineering and computer science research and applies it to real-world problems.

23 November 2020 - 27 November 2020

EECS Research Week 2020 is an exciting opportunity for our PhD students and academics to showcase their innovative and groundbreaking research. Discover our research in a series of online webinars.

In each webinar, PhD students and academic staff will showcase their research followed by a Q+A session and discussion between students considering applying for a PhD, researchers and industry.

Ten webinars will be delivered throughout the week - please register for as many as you wish to join! 

Click on tabs below to find out more about each webinar and follow the link to register.

Register Now!

Follow this link to register for the EECS Research Week 2020 webinars

Click here to register

23 November 2020, 11am - 12pm - Antenna Group

The Antennas & Electromagnetics Research group was established in 1968 with a mission of Image of antennas research lab“High Quality Research backed by High Quality Measurements”. The group currently has a strong team of academics and researchers working on various areas related to antenna engineering, bioelectromagnetic, novel materials for enhanced performance, antenna and electromagnetics (EM) theory and metrology concepts. Interdisciplinary research interfacing with materials, physics and chemistry, life sciences and medicine has led to new transformative research ideas.

The group has strong links with industry and its work has raised the international status of UK antenna research attracting many citations and a number of patents. Over its 50 year history the group has built an extensive Antenna Measurement Laboratory including antenna test ranges and electromagnetic characterisation facilities, which now covers the whole RF to THz frequency bands, from 400 MHz to 3 THz. Supported by three full-time professional staff, the facility is one of the most comprehensive among European universities and is frequently hired by industry: http://antennas.eecs.qmul.ac.uk/facilities/antenna-measurement-laboratory/.

Recently the group has doubled its size to 14 academic staff, extending its portfolio of research to: THz spectroscopy for space exploration; power electronics and battery control technology for Green Energy; digital manufacturing of antennas via 3D printing; 5G and beyond wireless communications.

Programme

Dr Akram Alomainy - “Multi-scale EM Solutions for Biomedical and Healthcare Applications”

Dr James Kelly - “Reconfigurable antennas based on liquid metal”

Dr Rob Donnan - "Doing Biophysics with Antenna Transceiver Systems"

Dr Rosty Dubrovka -  "Enhancing Harmonic Bee Tracking Radar Performance"

Professor Xiaodong Chen – “A Large Compact Antenna Test Range in Millimetre Wave and THz Bands”

Dr Henry Giddens - “3D printed dielectric antennas from microwaves to THz”

Dr Zhijin Qin - “Semantic Communication Systems”

Dr Kamyar Mehran - “Multi-physics EM modelling for the reliability of e-mobility power electronics”

Dr Sergio Ioppolo – “Ice Space Science at FELIX Laboratory”

Dr Flynn Castles - "THz Antenna Fabrication and Measurement Facilities (TERRA)"

23 November 2020, 11am - 12pm - Risk Information Management Group

Logo for the RIM Group

This webinar, presented by the Risk and Information management group, will focus on decision-support under uncertainty using methods from computer science, statistics, machine learning and psychology, to solve problems and challenges presented by scale, complexity, and variability. The research of the group is world-leading in its unique combination of data centric methods with hypothesis-driven approaches in which the power of advanced computing is combined with the insights of human expert judgements through Bayesian networks. The presentations will cover interdisciplinary topics ranging from chronic medical condition management to product safety assessment, as well as methods for learning the structure of causal models.

Bayesian networks are composed of directed graphs that represent variable dependencies, and quantitative probability assignments. These models are developed with practitioners to produce intelligent ‘unified models’ that use both data and expertise as inputs. They are then used to support inference and decision making in a wide range of application domains, including medical, legal, systems engineering, bioinformatics, security, risk and safety. 

Programme

  1. Dr Mariana Neves Using Bayesian Networks for Gestational Diabetes management support 
  2. Joshua Hunte (PhD Student) A novel method of product safety and risk assessment 
  3. Yang Liu (PhD Student)- Bayesian Network structure learning under the presence of measurement error 
  4. Kiattikun Chobtham (PhD Student) The White-Box Machine Learning: Bayesian Network Structure Discovery with Latent variables 
  5. Mohammad Bahrani (PhD Student) Towards Balanced and Generalizable Concept-based Models for Effective Medical Ranking 
  6. Tuomas Ketola (PhD Student) - Information Content-based Field Weighting for BM25F 
  7. Zhigao Guo (PhD Student) - Approximate Structure Learning of High-dimensional Bayesian Networks 
  8. Morghan Hartmann (PhD Student) - Assessing Multiple Sclerosis Risk Factors with Bayesian networks 
  9. Ali Fahmi (PhD Student) - Bayesian Networks for Managing Rheumatoid Arthritis 
  10. Dr Chris Joyner- TBC 

24 November 2020, 11am - 12pm - Game AI Research Group

Image of a computer game

The Game AI Research Group was founded in 2017 and has already comprises 10 academic staff and 25 PhD students and post-docs.  We use games both as a test-bed for the most challenging problems in AI, and as an important application game.  Much of our AI is based on very general statistical forward planning algorithms such as Rolling Horizon Evolution and Monte Carlo Tree Search.  We apply this to a range of challenging games, both for bot AI and to automatically generate and blend novel games and game content.  We work closely with a range of industry and government partners, and increasingly apply our AI to more complex games and real-world decision-making problems.  We are proud to co-host the IGGI (Intelligent Games and Game Intelligence) CDT, training more than 120 PhD students in games and Game AI.

As a group we contribute strongly to the Game AI research community, and play major roles in the main conferences such as IEEE Conference on Games, Foundation of Digital Games, and AI and Interactive Digital Entertainment (AIIDE), as well as the leading journal in the area, the IEEE Transactions on Games.

Welcome to our webinar, where you’ll find 10 talks that give a rough idea of what we do: read our papers and talk to us to find out more!

  1. Dr Paulo Rauber
  2. Raluca Gaina (PhD Student)
  3. Professor Simon Colton
  4. Dr Mike Cook
  5. Dr Diego Perez-Liebana
  6. Alvaro Ovalle (PhD Student)
  7. Dr Vanessa Volz
  8. Ivan Bravi (PhD Student)
  9. Professor Simon Lucas
  10. TBC

Programme

1. Dr Paulo Rauber

Title: Recurrent Neural-Linear Posterior Sampling for Non-Stationary Contextual Bandits

Abstract: An agent in a non-stationary contextual bandit problem should balance between exploration and the exploitation of (periodic or structured) patterns present in its previous experiences. Handcrafting an appropriate historical context is an attractive alternative to transform a non-stationary problem into a stationary problem that can be solved efficiently. However, even a carefully designed historical context may introduce spurious relationships or lack a convenient representation of crucial information. In order to address these issues, we propose an approach that learns to represent the relevant context for a decision based solely on the raw history of interactions between the agent and the environment. This approach relies on a combination of features extracted by recurrent neural networks with a contextual linear bandit algorithm based on posterior sampling. Our experiments on a diverse selection of contextual and non-contextual non-stationary problems show that our recurrent approach consistently outperforms its feedforward counterpart, which requires handcrafted historical contexts, while being more widely applicable than conventional non-stationary bandit algorithms.

2. Raluca Gaina (PhD Student)

Title: Artificial Intelligence in Modern Tabletop Games

Abstract: Tabletop games come in a variety of forms, including board, card, role-playing and dice games. In recent years, their complexity has increased considerably, with many components, rules that change throughout the game, diverse player roles, and many other aspects that can influence a game's balance. As a result, modern tabletop games also create new and interesting challenges for Artificial Intelligence methods, yet current research largely focuses on classical board games, such as chess and Go. This talk introduces the Tabletop Games (TAG) project, which instead looks into adaptive and immediately applicable general AI in modern tabletop games. The TAG system already includes a subset of games such as Colt Express, Exploding Kittens and Pandemic, as well as AI players capable of playing them all to some degree of proficiency. It also allows for easy development of new games and artificial players, complete with analytics that capture the complexities of the challenges proposed.

3. Professor Simon Colton

Title: Casual Creators Apps for Fun Creativity Support

Abstract: It's been possible for decades for people to use computers to make interesting art, music and games.

However, only in recent years has attention been paid to how to design creativity support apps that are - above anything else - fun to use. In the talk, I will describe some design and technical issues that arise in building such casual creators for general use, drawing on my experiences in developing the Wevva game design app and the Art Done Quick casual creator for visual art.

4. Dr Mike Cook

 Title: TBC

5. Dr Diego Perez-Liebana

Title: General Strategy Games Framework

Abstract: Strategy games are complex environments often used in AI-research to evaluate new algorithms. Despite the commonalities of most strategy games, often research is focused on one game only, which may lead to bias or overfitting to a particular environment. In this talk, we motivate and present a general strategy games framework for playing n-player turn-based and real-time strategy games. Our benchmark, which allows an easy customization of games via YAML-files, exposes an API with access to a forward model to facilitate research on statistical forward planning agents, as well as logging functionality for analyzing, debugging and understanding how algorithms reason in these complex environments.

6. Alvaro Ovalle (PhD Student) 

Title: Open issues in learning and planning with forward models

Abstract: Having a predictive forward model of the world confers an agent multiple advantages such as better generalization or the possibility of anticipating future consequences. However, acquiring and then planning over a learned model also comes with many challenges. In this talk, we present recent research trying to tackle two of them. The first case study considers a strategy for planning with inaccurate models. Via ensemble methods, the agent aggregates and performs error-correction on its predictions leading to more robust behavior. The second example deals with the scenario of planning without reward functions. We present a self-supervised planning approach based on active inference, which allows us to replace the notion of reward as utility, with that of maximizing model evidence in belief-spaces.

7. Dr Vanessa Volz

Title: Data-Driven Representations for Game Level Evaluation

Abstract: Game levels are complex artifacts, and different players enjoy them for different reasons. Some might enjoy puzzle-solving aspects, others having to prove their quick reaction skills. In some cases, however, even the player cannot explain the reasons for their enjoyment. Learning representations with data-driven methods is one approach to automatically characterise levels, that also allows for comparisons between them. In this talk, I will present a vision of how data-driven representations for game levels can help evaluate and compare them.

8. Ivan Bravi (PhD Student) 

Title: You better behave! - A general approach to behavioural exploration in games

Abstract: Evaluating the design of a game is a very complex task, playtesting is the primary tool for getting feedback from the players. This process is expensive and time consuming, grouping and analysing the feedback is complex and not free from bias. Using AI has the potential of speeding up the process, providing an unbiased and thorough quantitative analysis. This talk will present how to improve the ability of AI player to explore the behavioural space of a game.

9. Professor Simon Lucas

Title: Hierarchical Planning in Games

Abstract: Statistical Forward Planning algorithms such as Monte Carlo Tree Search and Rolling Horizon Evolution often perform amazingly well across a range of games, and can be further boosted by learning value functions or heuristics to help guide the search.  However, in some cases the action-space of a game is low-level and requires long sequences on order to achieve meaningful effects, causing particular difficulties when the reward landscape is flat.  In these cases a solution is to form plans in a higher level or macro action space.  In this talk I’ll present our latest work on automatically constructing macro-actions using a variety of state observation filters, subgoal predicates, and search procedures.

 

25 November 2020, 11am - 12pm - Advanced Robotics @ Queen Mary (ARQ)

The Centre for Advanced Robotics @ Queen Mary (ARQ) is a cross-faculty multidisciplinaryImages of robotic projects at qmul research centre which brings under one roof the different robotics-related activities performed at the Queen Mary University of London (QMUL). The main objective of ARQ is to facilitate collaborations within QMUL to reach critical mass and maximise societal impact. While interacting with different stakeholders from the industry and from the society at large, the members of ARQ provide world-class research and deliver advanced teaching programmes to both undergraduate and postgraduate students. The Centre is home to more than 50 researchers, including academics, research assistants and PhD students.  

 
In this webinar, you will learn more about our research activities in the areas of soft robotics, human-robot interaction, robotic solutions for hazardous environments, robots for industrial manufacturing, space robotics, robotic locomotion, assistive robotics, cognitive robotics, artificial intelligence, tactile sensing, haptics, Virtual Reality and wearable technologies. 

Programme

  • Professor Kaspar Althoefer, Head of Centre for Advanced Robotics, (EECS)
    • Perception in soft robots
    • Eversion robots to the rescue: new approaches to handle disaster situations  
    • Inflatable robot links: a solution for human-robot cooperation in industry
  • Dr Lorenzo Jamone (EECS)
    • Cognitive Robotics: understanding humans to create the robots of the future 
    • Using magnetism and soft materials to provide robots with a sense of touch  
  • Dr Ildar Farkatdinov (EECS)
    • - Wearable robots to support mobility
    • Haptic interfaces for human-machine interaction  
    • Robotics and virtual reality
  • Dr Angadh Nanjangud (SEMS)
    • Space robotics technologies for constructing large space structures
  • Dr Ketao Zhang (SEMS)
    • Legged mobile robots: leveraging mechanical intelligence for high-speed locomotion

25 November 2020, 11am - 1.15pm - Multimedia and Computer Vision Groups

Vision group Image

Multimedia and Vision: MMV was founded in 2000 and over the past 15 years, the group has made substantial contributions in several fields of multimedia signal processing including video compression, visual information retrieval and video analytics for security applications. 

The Queen Mary Computer Vision Group has been conducting world-leading research in computer vision and machine learning for over 20 years, and is internationally renowned for its work on video behaviour and action recognition, person re-identification, multi-camera tracking, and face analysis.

This webinar will see presentations from members of both groups exploring their innovative research into machine learning, image and video processing, person and action recognition, medical image analysis and audio-visual signal processing. 

Programme

Session 1 (30mins presentations) 

Machine Learning 

  • Woody Bayliss (PhD Student) - "Machine Learning Interpretability” 
  • Chen Feng (PhD Student) - “Self-supervised learning of visual representations 
  • Raymond Huang (PhD Student) - “Deep Semantic Clustering by Partition Confidence Maximisation” (CVPR 2020) 
  • Pan Li (PhD Student) - “Optimising Pseudo Labels for Semi-Supervised Few-Shot Classification” 
  • Ali Shahin Shamsabadi (PhD Student) - “Colorfool: Semantic adversarial colorization” (CVPR 2020) 
  • Dr Changjae Oh - ““Edge-fool: An adversarial image enhancement filter” (ICASSP 2020) 

Image and Video Processing 

  • Issa Khalifeh (PhD Student) - "Deep Learning for Video Temporal Super-Resolution" 
  • Bingqing Guo (PhD Student) - "Analysis of painting styles with deep-features and style innovation" 
  • Janice Li (PhD Student) - “Cast-GAN: Learning to remove colour cast from underwater images” 
  • Alessio Xompero (PhD Student) - “Multi-view shape estimation of transparent containers” (ICASSP 2020) 

Q&As (30mins) 

Coffee Break (15mins) 

Session 2 (30mins presentations) 

Person and Action Recognition 

  • Gavin Wu (PhD Student) "Tracklet Self-Supervised Learning for Unsupervised Person Re-Identification" (AAAI 2020) 
  • Samadhi Wickrama Arachchilage (PhD Student) - “Self-supervised Domain Learning for Face Recognition in the Wild" 
  • Georgios Zoumpourlis (PhD Student) - "Few-Shot and Zero-Shot Action Recognition using Temporal Attentive Relation Networks" 
  • Tingting Xie (PhD Student) - "Action Localisation with Uncertainty Aware Networks" 

Medical Image Analysis 

  • Xingru Huang (PhD Student) - "Lumen and Media Border Detection in Intravascular Ultrasound Sequences Using Attention Unet" 
  • Yibao Sun (PhD Student) - "Signet Ring Cells Detection in Histology Images with Similarity Learning" 
  • Yilong Li (PhD Student) - "SU-Net and DU-Net Fusion for Tumour Segmentation in Histopathology Images" 
  • Niki Foteinopoulou (PhD Student) - "Vision-based recognition of schizophrenia symptoms" 

Audio-Visual Signal Processing 

  • Dr Lin Wang - “Audio-visual sensing from drones 
  • Ashish Alex (PhD Student) - “Deep-learning for blind speech separation 

Q&As (30mins) 

25 November 2020, 11am - 12pm - Theory Group

Queen Mary is a world-leading centre for research on logical methods for reasoning about computer systems. Our work has spearheaded several developments – separation logic, logic for continuous systems, information theory for security, process types for web services, game semantics for programming languages – in which novel theoretical developments by us have been brought to bear in new application areas. We have also made fundamental contributions in pure logic (model theory, proof theory, categorical semantics) and in complexity theory.

In the last few years we have been awarded about £8m in research funding, supporting a thriving intellectual community. This includes an EPSRC “platform grant” (awarded to leading research groups in the UK to underpin their strategic development), and two EPSRC “programme grants” (a flexible mechanism for providing funding to address significant major research challenges, in research programmes of up to six years).

Programme

TBC

26 November 2020, 11am - 12pm - Centre for Digital Music

The Centre for Digital Music is a world-leading multidisciplinary research group in the field of Music & Audio Technology. Since its founding members joined Queen Mary in 2001, the Centre has grown to become arguably the UK’s leading Digital Music research group.

Programme

Title: Listen like a bat (machine learning & bat echolocation)

Presenter: Karol Bakunowski (PhD Student)

 

Title: Listen like a bird (machine learning & bird perceptual studies)

Presenter: Veronica Morfi (PhD Student)

 

Title: Building intelligent musical interfaces for extended reality

Presenter: Dr Mathieu Barthet

 

Title: Musical smart city (AI & sonification to make sense of IoT data)

Presenter: Pedro Sarmento (PhD Student)

 

Title: Automatic Lyrics Transcription

Presenter: Emir Demirel (PhD Student)

 

Title: Optical Music Recognition 

Presenter: Elona Shatri (PhD Student)

 

Title: Intelligent Audio Effects (or Audio Commons: open sound retrieval)

Presenter: Dr George Fazekas

 

Title: Hearing Loss Simulation

Presenter: Angeliki Mourgela (PhD Student)

 

Title: Introducing our New Lab!

Presenter: Dr Charis Saitis

 

Title: Auditory Overviews of Routes for Pre-navigation

Presenter: Dr Tony Stockman

26 November 2020, 11am - 12pm - Communications Systems Research (CSR) Group

LOGO for CSR research group

Welcome to the Webinar of Communication Systems Research (CSR) Group at QMUL. Our research spans the broad areas of wireless communications and statistical signal processing, with special emphasis on communication theory, information theory, optimization, machine learning, control theory, and random graph theory. The Communication Systems Research (CSR) group is internationally renowned for its contributions towards Fifth Generation (5G) Networks, Internet of Things (IoT), and Bio-inspired Molecular Communications.

There will be ten presentations on most recent beyond 5G research works at CSR. These presentations are in the areas of Artificial Intelligence (AI) enabled communications, Intelligent reflecting surface (IRS) enabled communications, internet of things (IoT) and Unmanned Aerial Vehicle (UAV) Networks.

Programme

1. Title: A Framework of Robust Transmission Design for IRS-aided MISO Communications with Imperfect Cascaded Channels

Authors: Gui Zhou (PhD Student); Dr Cunhua Pan; Hong Ren (PhD Student); Kezhi Wang; Professor Arumugam Nallanathan, Head of the Communications Systems Research Group (HoG)

Presenter: Gui Zhou

2. Title: A Wireless-Vision Dataset for Privacy Preserving Human Activity Detection

Authors: Yanling Hao (PhD Student), Zhiyuan Shi, Dr Yuanwei Liu

Name of presenter: Yanling Hao (PhD Student)

3. Title: Resource Allocation in Uplink NOMA-IoT Networks: A Reinforcement-Learning Approach

Authors: Waleed Ahsan (PhD Student), Wenqiang Yi (PhD Student), Dr Zhijin Qin, Dr Yuanwei Liu, Professor Arumugam Nallanathan (HoG)

Name of presenter: Waleed Ahsan (PhD Student)

4. Title: Multi-Agent Reinforcement Learning in NOMA-aided UAV Networks for Cellular Offloading

Authors: Ruikang Zhong (PhD Student), Xiao Liu (PhD Student), Dr Yuanwei Liu, Professor Yue Chen

Name of presenter: Ruikang Zhong (PhD Student)

5. Title: A Novel Physics-based Channel Model for Reconfigurable Intelligent Surface-assisted Multi-user Communication Systems

Authors: Jiaqi Xu (PhD Student), Dr Yuanwei Liu

Name of presenter: Jiaqi Xu (PhD Student)

6. Title: Artificial Intelligence Empowered Unmanned Aerial Vehicle Networks

Presenter: Xiao Liu (PhD Student), Professor Yue Chen

7. Title: Unsupervised Learning Clustering and Dynamic Transmission Scheduling for Efficient Dense LoRaWAN Networks

Presenter: Mohammed Alenezi (PhD Student), Dr Michael Chai

8. Title: "IRS-aided Communication Systems with Imperfect Channel Acquisition"

Authors: Yasaman Omid (PhD Student); Dr Cunhua Pan; Dr Yansha Deng; Professor Arumugam Nallanathan (HoG)

Presenter: Yasaman Omid (PhD Student)

9. Title: ‘Analyzing Grant-Free Access for URLLC Service’

Authors: Yan Liu (PhD Student); Dr Yansha Deng; Dr Maged Elkashlan; Professor Arumugam Nallanathan (HoG)

Presenter: Yan Liu (PhD Student)

10. Title: A Reinforcement Learning Approach for Wireless Backhaul Spectrum Sharing in IoE HetNets.

Authors: Dr Mona Jaber and Dr Atm Shafiul Alam

Presenter: Dr Mona Jab

26 November 2020, 11am - 12pm - Networks Research Group

Research project for the Networks Research Group

The Networks Research Group was established in 1987 and is active in key areas of networking including Internet measurements, Software-defined Networking, Network programmability, Web systems, Topology theory and resilience, Distributed computing, and Social computing. The group has an international reputation for excellence; work is regularly published in prestigious venues such as ACM SIGCOMM, IEEE INFOCOM, ACM SIGCOMM IMC, ACM SIGCOMM CoNEXT, WWW, IEEE ICNP and the top journals in the field (e.g. IEEE/ACM ToN, IEEE TPDS, IEEE TC, ToMM, ACM SIGCOMM CCR).

The group has received major funding from EPSRC, EU H2020, and industrial partners.  The group is also highly active in community activities, recently organising conferences such as the Internet Measurement Conference in 2017, the flagship SIGCOMM conference in 2015, the symposium of Computing for Development in 2015 and the International workshop on Traffic and Monitoring and Analysis in 2014. Network group members serve as editors for lead publications in the field such as IEEE/ACM Transactions on Networking, ACM SIGCOMM CCR, and Computer Communications journal, and as general and TPC chairs of leading conferences in the field (e.g., ACM SIGCOMM, IEEE ICNP, IMC, PAM). Prof. Uhlig who leads the group is Editor in Chief of the ACM SIGCOMM CCR journal, the newsletter published by ACM SIGCOMM.

Programme

1) Title: Big data temporal graph systems with Raphtory 

Presenter: Dr Richard Clegg 

2) Title: Network emulation for precise QoE evaluation

Presenters: Nafi Ahmed (PhD Student), Abdul Wahab (PhD Student)

3) Title: Programming the network

Presenter: Dr Gianni Antichi

4) Title: Understanding how networks evolve (FETA)

Presenter: Naomi Arnold (PhD Student)

5) Tile: ML-enabled mobile network performance management based on streaming CDR data

Presenter: Dr Mona Jaber

6) Title: Simplification of networks via their path diversity

Presenters: Hengda Yin and Dr Raul Mondragon 

7) Title: Learning resource management in cloud computing

Presenter: Saeid Ghafouri (PhD Student)

27 November 2020, 10am - 11pm - Cognitive Science

 

CogSci at QMUL is a vibrant community of researchersthe Cover of EECS produced magazine - CS4FN. Here, we study human cognition, action and interaction on scales ranging from individual experience, through interactions between individuals, to the languages, cultures and dynamics of societies. With principal labs including computational linguistics, education, human interaction, music cognition, wearables and social science projects tackle important and complex topics. We collaborate not only with other QMUL centres such as digital music, mind in society, psychology, life sciences and linguistics but also with local, national and international bodies. Attracting funding from government, research consortia and industry we run local conferences, international seminars and large scale research programmes.

In just 30 minutes we will spotlight 10 of our current Cogsci programmes. From robots making sense of a noisy world to monitoring mental health through language, come and find out more about our ground-breaking work.

The Education Lab in the CogSci have been producing cs4fn, a magazine about computer science. In our lastest issue we celebrated the life of Peter McOwan, an inspirational founder of the magazine and principle CogSci researcher. Come along to our online session and we will send you a copy!

 

Programme

1. Professor Paul Curson and Jane Waite (Teaching Fellow & Public Engagement, Outreach and Teacher CPD Co-ordinator):  
Title: Education Lab:  Effective learning in CS and EE 
Short Description:  As a relatively new subject there is only limited understanding of the best ways to teach computer science and electronic engineering especially at school level whether with or without educational technology support. Our research explores, for example, the effectiveness of unplugged computing, the importance of design in teaching programming, and ways to teach computational thinking. A particular lens we are using is called semantic waves, a way to think about what makes a good explanation of concepts.   

2. Dr Julian Hough:  
Title: Making Human-Robot Dialogue Fast and Fun 
Short Description: While Human-Robot Interaction (HRI) is a blossoming young field, relatively little work has been done in improving interaction with robots with speech understanding capacities from a quality of experience point of view. Our research focusses on improving the smoothness of HRI by improving the speed of processing speech as it is received, reducing lengthy delays by robots, and processing signals of misunderstanding in a natural way, given the inherent noisiness of the real world as the robot senses it. 

3. Professor Patrick Healey, Head of the Cognitive Science Research Group:   
Title:  Social Health 
Short Description:  Conversation has a profound impact on health and wellbeing. Technology provides opportunities to map individual and communal the social health and give people the means to extend and enrich their opportunities for interaction.  

4. Dr Lorenzo Jamone:  
Title:  Cognitive Robotics 
Short Description: Creating intelligent robots by taking inspiration from humans, and using robotics research to better understand human intelligence. 

5. Professor Maria Liakata/ Dr Adam Tsakalidis  
Title:  Towards time sensitive sensors from language and heterogeneous user content 
Short Description:  The majority of work in NLP on mental health prediction, even when using longitudinal social media data, involves distinguishing individuals with a condition from controls rather than assessing individuals’ mental health at different points in time. This talk will present our goals to create such time-sensitive assessments for individuals and progress we have so far. 

6. Dr Marcus Pearce:  
Title:  Music Cognition 
Short Description: Like spoken language, music is universal across cultures and involves the creation of sequences of sounds which are interpreted by suitably enculturated listeners. Research in music cognition aims to understand the psychological and neural processes involved in the production, perception and appreciation of music, how these are acquired and their relationships with other psychological phenomena.    

7. Professor Massimo Poesio:  
Title:  Games and NLP (?).  
Short Description: Studying language with computational means requires collecting from people millions of judgments about the way they use language, and engaging people via games has proven a great way to crowdsource such judgments on a large scale. Research on Games With A Purpose is concerned with developing games that are enjoyable yet can collect data of scientific interest. 

8. Professor Matthew Purver: 
Title:  Monitoring Progress in Therapy Through Language 
Short Description:  The language people use can tell us a lot about their mental states and their mental health, via both the content of what they say and how they interact with others. This talk explains work joint with IESO Digital Health into how we can use computational natural language processing (NLP) techniques to help therapists monitor their patients’ progress and predict therapy outcomes. 

9. Professor Matthew Purver 
Title:  Automatic News Comment Moderation Across Languages 
Short Description:  Online newspapers usually allow their readers to comment on the stories they publish, but this comes with a problem: comments must be monitored to check for anything off-topic, abusive or illegal, and it’s hard to keep up with the volume of data if you have to do it manually. Automatic tools, on the other hand, are hard to develop, and only exist for major languages like English. This talk will explain work joint with news publishers around Europe and show how we can quickly develop tools that work in less-resourced languages without needing existing datasets or relying on machine translation. 

10. Dr Laurissa Tokarchuk:  
Title:  Immersive Experiences in Mixed Reality Games 
Short Description: This talk will explore immersion and augmented reality location based games. Specifically, it will explore tools and techniques for using player movement, difficulty adjustment, location and graphical overlay mapping to provide immersive games in augmented reality 
    
11. Dr Arkaitz Zubiaga:  
Title:  Social Data Science 
Short Description: Social data science is concerned with the use of a wide range of computational and data science methods to analyse social phenomena through (often large-scale) datasets.