Author: Benjamin

Posted on

IC8 – Computational principles of gaze-stabilization during locomotion

Lecturer: Hans Straka
Fields: Experimental Neurobiology

Content

Continuous accurate perception of the visual world is a behavioral requirement during self-generated motion. All animals are confronted with the disruptive effects of locomotor activity on the ability to maintain stable images on the retina. This is due to the fact that self-motion is accompanied by head movements that cause retinal image displacement with a resultant degradation of visual information processing. To stabilize gaze and to retain visual acuity during locomotion, retinal image drift is offset by counteractive eye and/or head-adjustments. These offsetting motor reactions are classically attributed to the concerted action of visuo-vestibular and proprioceptive reflexes. However, stereotyped, rhythmic locomotion has predictable consequences for image perturbations. This, in principle, allows employing efference copies of propulsive motor commands to directly initiate spatio-temporally adequate eye movements. Such eye-adjusting motor commands have been demonstrated in the amphibian Xenopus laevis. These signals are feed-forward replica of the spinal central pattern generator output that produces the actual propulsive body movements. Spinal locomotor efference copies directly target horizontal extraocular motoneurons, consistent with the plane and direction of swimming-related head rotations. The signals actively attenuate vestibulo-ocular reflexes, emphasizing the predominant role for intrinsic efference copies for gaze-stabilization during self-motion. The suppressive influence of motor efference copies on vestibular signals occurs at the mechanosensory periphery. The resultant gain reduction in sensory signal encoding likely prevents overstimulation by adjusting the system to increased stimulus magnitudes during locomotion. This leaves efference copy-evoked gaze-stabilizing eye movements as dominant computational mechanism. Further suggestive evidence for a ubiquitous role of such signals in this context has been provided for quadrupedal and bipedal locomotion in terrestrial vertebrates including humans.

Literature

  • Lambert F.M., Combes D., Simmers J. and Straka H. (2012) Gaze stabilization by efference copy signaling without sensory feedback during vertebrate locomotion. Curr. Biol. 22: 1649-1658.
  • Chagnaud B.P., Simmers J. and Straka H. (2012) Predictability of visual perturbation during locomotion: implications for corrective efference copy signaling. Biol. Cybern. 106: 669-679.
  • von Uckermann G., Le Ray D., Combes D., Straka H. and Simmers J. (2013) Spinal efference copy signaling and gaze stabilization during locomotion in juvenile Xenopus frogs. J. Neurosci. 33: 4253-4264.
  • Chagnaud B.P., Banchi R., Simmers J. and Straka H. (2015) Spinal corollary discharge modulates motion sensing during vertebrate locomotion. Nat. Comm. 6: 7982 doi: 10.1038/ncomms8982.
  • von Uckermann G., Lambert F.M., Combes D., Straka H. and Simmers J. (2016) Adaptive plasticity of retinal image stabilization during locomotion in developing Xenopus. J. Exp. Biol. 219: 1110-1121.
  • Straka H., Simmers J. and Chagnaud B.P. (2018) A new perspective on predictive motor signaling. Curr. Biol. 28: R232-R243.

Lecturer

Hans Straka

Hans Straka is Professor for Systemic Neurosciences at the Faculty of Biology at the LMU Munich. He studied Biology at the LMU Munich and received his PhD from the same University. Starting with his postdoc, he got interested in the functional organization of the vestibular system including its variable morphology as well as the ontogeny and phylogeny of this sensory system. Using a variety of animal models, he has studied over the past years in the US, in France and currently in Munich the respective contributions of cellular and neural networks to the sensory transformation of head/body motion-related signals into appropriate extraocular motor commands. Interactions with computational neuroscientists have resulted in a number of conceptual novelties on gaze control and computational models that bridge the gap between empiric experiments and theoretical background.

Affiliation: Depatment Biology II, Ludwig-Maximilians-Universiy Munich
Homepage: https://neuro.bio.lmu.de/members/systems_neuro_straka/straka_h/index.html

Posted on

IC13 – Personalizing instruction and recognizing student misunderstandings using reinforcement learning

Lecturer: Anna Rafferty
Fields: Artifical Intelligence/Machine learning

Content

Online educational technologies provide opportunities to monitor learners’ knowledge in real time and modify instruction based on learners’ responses. In this talk, I’ll give a brief overview of some of the ways that reinforcement learning has been used to achieve these goals, and then provide a more in-depth discussion of my own work using inverse reinforcement learning to make inferences about learners’ understanding. In this work, we are particularly focused on interpreting learners’ behavior in multi-step tasks, such as games or mathematical problem solving, and we combine ideas from machine learning and computational cognitive modeling. Our approach offers the potential to provide feedback about learners’ strategies and misunderstandings based on their pattern of interactions. Overall, the talk will argue that work in reinforcement learning for education has the potential to create smarter educational resources and that taking an interdisciplinary perspective suggests new insights and approaches.

Literature

  • Rafferty, A. N., Jansen R. A., & Griffiths, T. L. (2020). Assessing Mathematics Misunderstandings via Bayesian Inverse Planning. Cognitive Science. DOI: 10.1111/cogs.12900
  • Rafferty, A. N., Jansen, R. A., & Griffiths, T. L. (2016) Using Inverse Planning for Personalized Feedback. Proceedings of the 9th International Conference on Educational Data Mining (pp. 472-477). http://tiny.cc/IRLFeedbackEDM2016

Lecturer

Anna Rafferty

Dr. Anna Rafferty earned her PhD from the University of California, Berkeley, and is currently an associate professor of computer science at Carleton College. Her work addresses questions at the intersection of machine learning, computational cognitive science, and education. She is particularly interested in developing automated strategies to provide effective feedback to students and in developing technologies that can both continuously improve instruction for students and provide valuable data for researchers to draw more general conclusions about the effectiveness of educational interventions. Dr. Rafferty has recently begun work emphasizing the importance of considering equitable impacts across students in educational technologies that have the potential for personalization.

Affiliation: Carleton College
Homepage: https://sites.google.com/site/annanrafferty/

Posted on

IC2 – The gendered nature of gamer stereotypes and what we can do about it

Lecturer: Thekla Morgenroth
Fields: Psychology

Content

Female gamers are seen as atypical and often have their competence challenged in gaming spaces. We argue that this is partly driven by masculine gamer stereotypes and that exposure to female gamers has the potential to change them. We investigate the content of gamer stereotypes across two studies and find that they contain both negative aspects, such as lacking social skills, and positive aspects, such as being competent and agentic. Both studies demonstrate that gamer stereotypes are more similar to stereotypes of men and boys than those of women and girls. In Study 2 we further find evidence suggesting that exposure to a female gamer can change the negative association between female stereotypes and gamer stereotypes. We conclude that increasing the visibility of female gamers could potentially reduce the incompatibility between femininity and gaming and alleviate some of the issues female gamers currently face.

Literature

  • Blackburn, G., & Scharrer, E. (2019). Video game playing and beliefs about masculinity among male and female emerging adults. Sex Roles, 80(5-6), 310-324.
  • Paaßen, B., Morgenroth, T., & Stratemeyer, M. (2017). What is a true gamer? The male gamer stereotype and the marginalization of women in video game culture. Sex Roles, 76(7), 421-435.
  • Wasserman, J. A., & Rittenour, C. E. (2019). Who wants to play? Cueing perceived sex-based stereotypes of games. Computers in human behavior, 91, 252-262.

Lecturer

Thekla Morgenroth

Dr. Thekla Morgenroth received their PhD in Social and Organizational Psychology from the University of Exeter in 2015. Their research focuses on how and why people maintain social hierarchies with a specific focus on the barriers encountered by members of the LGBTQ+ community and women.

Affiliation: University of Exeter
Homepage: https://psychology.exeter.ac.uk/staff/profile/index.php?web_id=Thekla_Morgenroth

Posted on

IC5 – Exploring your own mind

Lecturer: Marieke van Vugt
Fields: cognitive science, contemplative science

Content

During this talk, I will introduce the science of mind-wandering, and connect this to the topic of mindfulness. In the study of mind-wandering we empirically test how subjective experience influences objective measures such as EEG, eye-tracking and behaviour. In contrast, in mindfulness, we explore our own minds. How can you explore your own mind, and become more familiar with mind-wandering from the inside?

Literature

  • Huijser et al. (2020). Captivated by thought: “Sticky” thinking leaves traces of perceptual decoupling in task-evoked pupil size. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0243532

Lecturer

Marieke van Vugt
Prof. van Vugt

Dr. van Vugt is an assistant professor at the University of Groningen in the Netherlands, working in the department of artificial intelligence. She obtained her PhD in model-based neuroscience from the University of Pennsylvania, then worked as a postdoc at Princeton University before moving to the University of Groningen. In her lab, she focuses on understanding the cognitive and neural mechanisms underlying decision making, mind-wandering and meditation by means of EEG, behavioural studies and computational modeling. In some slightly outside-the-box research, she also records the brain waves of Tibetan monks and dancers.

Affiliation: University of Groningen
Homepage: https://mkvanvugt.wordpress.com

Posted on

ET4 – Biosignal Processing for Human-Machine Interaction

Lecturer: Tanja Schultz
Fields:

Content

Human interaction is a complex process involving modalities such as speech,
gestures, motion, and brain activities emitting a wide range of biosignals, which can be captured by a broad panoply of sensors. The processing and interpretation of these biosignals offer an inside perspective on human physical and mental activities and thus complement the traditional way of observing human interaction from the outside. As recent years have seen major advances in sensor technologies integrated into ubiquitous devices, and in machine learning methods to process and learn from the resulting data, the time is right to use of the full range of biosignals to gain further insights into the process of human-machine interaction.

In my talk I will present ongoing research at the Cognitive Systems Lab (CSL), where we
explore interaction-related biosignals with the goal of advancing machine-mediated human
communication and human-machine interaction. Several applications will be described such as Silent Speech Interfaces that rely on articulatory muscle movement captured by
electromyography to recognize and synthesize silently produced speech, as well as Brain
Computer Interfaces that use brain activity captured by electrocorticography to recognize
speech (brain-to-text) and directly convert electrocortical signals into audible speech (brain-to-speech). I will also describe the recording, processing and automatic structuring of human everyday activities based on multimodal high-dimensional biosignals within the framework of EASE, a collaborative research center on cognition-enabled robotics. This work aims to establish an open-source biosignals corpus for investigations on how humans plan and execute interactions with the aim of facilitating robotic mastery of everyday activities.

Objectives

None

Literature

None

Lecturer

Dr. Tanja Schultz

Tanja Schultz received her diploma (1995) and doctoral degree (2000) in Informatics from University of Karlsruhe and completed her Masters degree (1989) in Mathematics, Sports, and Educational Science from Heidelberg University, Germany.
Dr. Schultz is the Professor for Cognitive Systems at the University of Bremen, Germany and adjunct Research Professor at the Language Technologies Institute of Carnegie Mellon, PA USA. Since 2007, she directs the Cognitive Systems Lab, where her research activities include multilingual speech recognition and the processing of biosignals for human-centered technologies and applications. Since 2019 she is the spokesperson of Bremen’s high-profile area “Minds, Media, Machines”. Dr. Schultz is an Associate Editor of ACM Transactions on Asian Language Information Processing and serves on the Editorial Board of Speech Communication. She was President and elected Board Member of ISCA, and a General Co-Chair of Interspeech 2006. She is a Fellow of ISCA and member of the European Academy of Sciences and Arts. Dr. Schultz was the recipient of several awards including the Alcatel Lucent Award for Technical Communication, the PLUX Wireless Biosignals Award, the Allen Newell Medal for Research Excellence, and received the Speech Communication Best paper awards in 2001 and 2015.  

Affiliation: University of Bremen

Posted on

PC4 – Curious Making, Taking Fabrication Risks and Crafting Rewards

Lecturer: Janis Meißner
Fields: Design, Human Computer Interaction

Content

This course is about getting hands-on curious with electronics and different crafts materials. Maker toolkits are a great way to get started with designing your own interactive sensor systems – but what if these designs could also integrate other (potentially more aesthetic) materials? E-textiles and paper circuits are good examples for how functional electronic systems can be recrafted with rewarding results. In principle, any every-day materials could be used with a bit of thinking outside the (tool)box. Let’s see what you will use to hack for your ideas!

Course Outline:

After a brief intro to microcontrollers and programming them with the Arduino IDE, participants will design their own simple input-output systems and gradually re-craft the hardware in innovative ways by using crafting materials such as for example paper, fabric and paperclips. Participants who seek a little extra-challenge are invited to work in small teams (2-4) to design an interactive artefact in this way that combines their respective research interests.

The course is structured as follows:

Session 1: Introduction to microcontrollers, off-the-shelf components and self-paced experimenting with the help of tutorials

Session 2: Designing an input-output system with off-the-shelf components. Starting to explore how ready-made components can be re-made with crafts materials.

Session 3-4: Recrafting your system design with craft materials of your choice. Don’t forget to present your inventions to your course mates so that everyone can applaud your creative hacking genius! 🙂

Objectives

  • Learning the basics of programming electronics with microcontrollers
  • Learning the basics of how a selection of sensors and actuators work
  • Exploring alternative approaches to electronics than using o
  • Unleashing your creative hacking skills

Literature

Perner-Wilson, H., Buechley, L. & Satomi, M. (2011) ‘Handcrafting textile interfaces from a kit-of-no-parts’, in Proceedings of the fifth international conference on Tangible, embedded, and embodied interaction – TEI ’11. New York, USA: ACM Press. p. 61. https://doi.org/10.1145/1935701.1935715

Posch, I. & Fitzpatrick, G. (2018) Integrating Textile Materials with Electronic Making. Proceedings of the Twelfth International Conference on Tangible, Embedded, and Embodied Interaction – TEI ’18. 158–165. https://doi.org/10.1145/3173225.3173255

Meissner, J.L., Strohmayer, A., Wright, P. & Fitzpatrick, G. (2018) ‘A Schnittmuster for Crafting Context-Sensitive Toolkits’, in Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems – CHI ’18. New York, New York, USA: ACM Press. https://doi.org/10.1145/3173574.3173725

Lecturer

Janis Lena Meißner

Janis Lena Meißner is a doctoral trainee in Digital Civics at Open Lab, Newcastle University, and co-founder of fempower.tech, a group of intersectional feminists who aim to raise awareness of feminist issues in Human Computer Interaction. As maker technologies give individuals an opportunity to develop their own objects and tools, Janis is interested in exploring ways that these technologies can empower different non-technical communities who lack access to infrastructures such as fablabs or makerspaces. In her research she has collaborated with groups as diverse as urban knitters, glass artists, quilting sex workers, makers with disabilities and members of a Men Shed interested in combining their woodworking skills with 3D-printing. Using a Participatory Action Research methodology and a portable makerspace for adapting tool(kit)s to the specific contexts of making, her aim is to develop a community-driven approach to Making that allows people to weave in pre-existing crafting skills into their use of digital maker technologies.

Affiliation: Newcastle University
Websites: https://fempower.tech/ https://openlab.ncl.ac.uk/people/janis-lena-meissner/ https://twitter.com/janislena

Posted on

IC4 – Introduction to Ethics in AI

Lecturer: Heike Felzmann
Fields: Ethics, AI

Content

The last few years have seen an explosion of societal uses of AI technologies, but at the same time widespread public scepticism and fear about their use have emerged. In response to these concerns, a wide range of guidance documents for good practice in AI have been published by professional and societal actors recently. Both as researchers in AI and as consumers of AI it is helpful to understand ethical concepts and concerns associated with the use of AI and to be familiar with some of these guidance documents, in order to be able to reflect carefully on their ethical and social meaning and the balance of their benefits and risks and adapt one’s practices accordingly.

This course provides a general introduction to emergent ethical issues in the field of AI. It will be suitable for anyone with an interest in reflecting on how AI impacts on contemporary life and society. Over the four sessions of the course we will introduce and reflect on ideas and practical applications related to the following topics:

  • Understanding privacy, consent and transparency
  • Automated decision-making, algorithmic biases, autonomous artificial agents and accountability for decisions by artificial agents
  • Assistance, surveillance, persuasion, and human replacement
  • Responsible design and implementation, trustworthiness, and AI for good

Objectives

The goal of the course is for participants to gain familiarity with core ethical concepts and concerns arising in the development and societal uses of AI, allowing participants to engage in a differentiated and informed manner with the societal debates on AI.

Literature

Eubanks, V. (2018). Automating inequality: How high-tech tools profile, police, and punish the poor. St. Martin’s Press. (on Google Books)

HLEG on AI (2019) Ethics Guidelines for Trustworthy AI, https://ec.europa.eu/futurium/en/ai-alliance-consultation 

Nissenbaum, H. (2019). Contextual Integrity Up and Down the Data Food Chain. Theoretical Inquiries in Law, 20(1), 221-256. http://www7.tau.ac.il/ojs/index.php/til/article/download/1614/1715 (ignore the abstract, which is much more obscure than the rest of the article! Contextual integrity is a useful theory of privacy.)

Zuboff, S. (2019) The Age of Surveillance Capitalism: The fight for a human future at the frontier of power. (Youtube interviews with Zuboff might be a good introduction.)

Lecturer

Heike Felzmann is a lecturer in Ethics in the School of History and Philosophy at NUI Galway, Ireland. She works on ethics in information technologies (especially on healthcare robots and AI), research ethics, and general health care ethics. She has been part of several European projects, including H2020 MARIO on a care robot for patients with dementia, H2020 ROCSAFE on robot supported incident response, COST 16116 on robotic exoskeletons, COST RANCARE on rationing in nursing care, ITN DISTINCT on technology use in dementia care, ERASMUS PROSPERO on education on social robots for social care, and was the chair of the COST Action CHIPME on innovations in genomics for health. She has also had extensive experience with research ethics governance and research ethics training. She teaches ethics widely across disciplines and is looking forward to meeting the interdisciplinary audience at the IK.

Website: http://www.nuigalway.ie/our-research/people/humanities/heikefelzmann/

Affiliation: NUI Galway

Posted on

FC13 – Hominum-ex-Machina: About Artificial and Real Intelligence

Lecturer: Markus Krause
Fields: Human Computer Interaction, Artificial Intelligence (actually: Advanced Statistical Analysis and Pattern Recognition), Human Computation

Content

Modern computational systems have amazing capabilities. They can detect a face or fingerprint in millions of samples, find a search term in a sea of billions of documents, and control the flow of trillions of dollars. Some of these abilities seem almost supernatural and even frightening. Yet, our brains are still the architects of invention and might remain to be so for aeons to come. Understanding and utilising the difference between machine und human intelligence is one of the new frontiers of computer science. With the advent of the next AI winter integrating human intervention into almost autonomous systems will gain crucial importance in the near future.

In this course we aim at lifting a bit of the mystic shroud that surrounds artificial intelligence. We will uncover its abilities, unveil short comings, and even conjure a deep neural network from (almost) thin air. You do not need to be an experienced coder or mathematical genius. Basic python understanding, and 8 grade math skills are enough to follow the course and build your own “AI”. After this hopefully disillusionary exercise we take a refreshing dive into reality. We will investigate real intelligence and how our brains talent for strategic problem solving can fuse with the sheer calculation power of machines. We will explore how these socio-technical systems will shape the future and the risks and pitfalls of the Hominum-ex-Machina.

Objectives

Understanding the limitations of machine-based decision capabilities, the abilities setting humans apart from computers, and how human and machine abilities can fuse to form large scale computational systems.

Literature

Interesting AI Papers:
David Saxton: https://arxiv.org/pdf/1904.01557.pdf
Rumelhart et al: Learning internal representations by error-propagation
Krizhevsky et al: Imagenet classification with deep convolutional neural networks
Hochreiter, Schmidhuber: Long short-term memory
A Vaswan et al: Attention is all you need

HComp Papers:
https://dl.acm.org/conference/chi
https://dl.acm.org/conference/cscw
https://www.aaai.org/Library/HCOMP/hcomp-library.php

First Paper about Human Computation and the inverse Turing test: http://www.wisdom.weizmann.ac.il/~naor/PAPERS/human.pdf

Book by Luis von Ahn and Edith Law basics about the inverted touring test at work: https://www.google.com/books/edition/Human_Computation/bF7ePcj-cUMC?hl=en&gbpv=1&printsec=frontcover

A set of interesting papers to take crowdsourcing to a higher complexity level:
https://hci.stanford.edu/publications/2017/flashorgs/flash-orgs-chi-2017.pdf
https://hci.stanford.edu/publications/2017/crowdresearch/crowd-research-uist2017.pdf
https://www.mooqita.org/publications/empoweringhiddentalents.pdf

Brian Christian’s account of participating in the Turing test yearly competition https://www.amazon.com/Most-Human-Artificial-Intelligence-Teaches/dp/0307476707

Lecturer

Dr. Markus Krause
Dr. Markus Krause

Dr. Markus Krause is a computer scientist, professional game designer, and serial entrepreneur. He co-founded Mooqita a Berkeley based Non-Profit supporting students in finding the job they love. Mooqita uses a novel approach combining human and machine intelligence. Dr. Krause also co-founded Brainworks.ai. Brainworks develops a new neural cortex to use smartphones as diagnostic tools for online health care applications. He is also the primary investigator for the Mooqita project at the International Computer Science Institute at UC Berkeley and part of the advisory committee to the DAAD IFI. Dr. Krause earned is doctoral degree in computer science from the University of Bremen, Germany and the Carnegie Mellon University in Pittsburgh, USA.

Websites: https://www.mooqita.org/
http://brainworks.ai/
https://www.linkedin.com/in/markus-krause-3490b246/