The following projects have received funding from Chalmers AI Research Centre.
Learning & Understanding Human-Centered Robotic Manipulation Strategies
Research Project, 2020–2025
PhD project
Human-robot collaboration is one of the most promising applications of autonomous robots. For this, robots need to interact with humans in a meaningful, flexible and adaptable manner, especially when new situations are faced. Currently, the new emerging technologies such as virtual reality and wearable devices allow capturing natural human movements of multiple users. Then, the next generation of learning methods should take advantage of and bootstrap the learning of new activities by adapting to the massive processing of information obtained from these enhanced sensors.
PI: Karinne Ramirez-Amaro, Electrical Engineering, Chalmers
Intelligent agents that learn from their past experiences
PhD project
Embodied Artificial Intelligence is a multidisciplinary area that requires the cooperation of different fields such as computer science, engineering, robotics and dynamical systems. This PhD thesis will develop a novel learning algorithm to allow high-level intelligence, such as problem-solving and reasoning to be applied to real-world physical systems, e.g. robots. Mainly, this work will be focused on investigating learning methods on the semantic aspects of intelligence to develop general purpose solutions for robotic applications.
PI: Karinne Ramirez-Amaro, Electrical Engineering, Chalmers
Dexterous robot assistant for everyday physical object manipulation
PostDoc project
The project will focus on robotic grasping and manipulation, a challenging sub-area of robotics, with a strong potential to have a large positive impact on society. The applications of robotic manipulation range from assistive robotics, taking care of elderly and disabled people, and helping humans with day-to-day tasks, to automating manufacturing and replacing humans at tedious and dangerous jobs. The project will build novel methodologies, advancing the state of the art, to realize a robust robotic system that can achieve complex manipulation tasks in high clutter and occlusions operating with complex objects, e.g. deformable, articulated, autonomously; extend its repertoire of such by acquiring new skills; encode knowledge from experience; build goal-oriented behaviours with dexterity and high level reasoning and planning capabilities; and re-plan the execution cycle to robustly cope with unexpected events and failures.
PI: Yasemin Bekiroglu, Electrical Engineering, Chalmers
Geometric Deep Learning
PostDoc project
The project aims to investigate how inherent geometrical properties of the data can be incorporated in deep learning methods. To be concrete, consider a supervised learning task. The data set having an ‘inherent geometrical property’ can mathematically expressed as the labels depending either invariantly or equivariantly on the data point with respect to some group action. Can this invariance/equivariance be guaranteed by a certain architecture design? Convolutional neural networks are in fact an example of such an architecture – if the input image is translated, the output of a convolutional layer is translated with it.
PI: Fredrik Kahl, Electrical Engineering, Chalmers
Information theory of deep neural networks
PhD project
Over the last decade, deep-learning algorithms have dramatically improved the state of the art in many machine-learning problems, including computer vision, speech recognition, natural language processing, and audio recognition. Despite their success, however, there is no satisfactory mathematical theory that explains the functioning of such algorithms. Indeed, a common critique is that deep-learning algorithms are often used as black box, which is unsatisfactory in all applications for which performance guarantees are critical (e.g., traffic-safety applications).
The purpose of this project is to increase our theoretical understanding of deep neural networks. This will be done by relying on tools of information theory—the area of expertise of the PI, Giuseppe Durisi—and focusing on specific tasks that are relevant to computer vision—the area of expertise of the co-PI, Fredrik Kahl.
PI: Giuseppe Durisi, Electrical Engineering, Chalmers
Verification of Machine Learning Algorithms
PostDoc Project
It has long been known that our ability to develop and deploy machine learning (ML) algorithms outpaces our ability to make clear guarantees about their behaviour. This situation is unacceptable, as ML algorithms will increasingly be deployed in safety critical systems, for example in autonomous vehicles. In this project, we aim to develop new methods of testing and verifying machine learning algorithms and to kickstart our group’s application of its expertise in testing and formal verification in the area of AI/ML.
PI: Mary Sheeran, Computer Science and Engineering, Chalmers
AI-Integromics: Intepretable Deep Modeling for Systems Biology
Postdoc/PhD project
We aim to provide AI solutions for Integromics that are effective and interpretable. To this end, an AI-Integromics approach should predict phenotypes for in-house data, attribute predictive strength to the multi-omics source data, yet also provide a mapping to extensive public data resources to facilitate drug re-purposing. We propose a complete AI-Integromics pipeline, constituting a novel framework that combines research on structured regularization of deep NNs and GANs with semi-parametric modeling in statistics.
PI: Rebecka Jörnsten, Mathematical Sciences, Chalmers
We construct new deep neuronal networks (DNNs) to learn the
parameters of complex stochastic dynamical models that do not have
tractable likelihood functions. Specifically, we leverage the expressive
approximation power of our DNNs to extract essential information from
time-series data, both Markovian and not-Markovian, and then learn model
parameters using likelihood-free methodology, such as approximate
Bayesian computation. Special (though not exclusive) focus is directed
to stochastic differential equation models and state space models
(SSMs), where SSMs represent noisy observations of a latent Markovian
process. The result will be a flexible plug-and-play machine learning methodology, allowing inference for complex stochastic models.
Applicant: Umberto Picchini, Department of Mathematical Sciences, Chalmers
Description missing.
Applicants: Aikaterini Mitrokotsa, Department of Computer Science and Engineering, Chalmers, Christos Dimitrakakis, Department of Computer Science and Engineering, Chalmers
Stochastic continuous-depth neural networks
Research Project, 2020 –We will advance the understanding of deep neural networks through the
investigation of stochastic continuous-depth neural networks. These can
be thought of as deep neural networks (DNN) composed of infinitely many
stochastic layers, where each single layer only brings about a gradual
change to the output of the preceding layers. We will analyse such
stochastic continuous-depth neural networks using tools from stochastic
calculus and Bayesian statistics. From that, we will derive practically
relevant and novel training algorithms for stochastic DNNs with the aim
to capture the uncertainty associated with the predictions of the
network.
Applicant: Moritz Schauer, Department of Mathematical Sciences
Understanding and Overcoming the Limitations of Convolutional Neural Networks for Visual Localization (VisLockLearn)
Research Project, 2020–2024Visual localization is the problem of estimating the position and
orientation from which an image was taken with respect to the scene. In
other words, visual localization allows an AI system to determine its
position in the world through a camera. Understanding why current
approaches fail and proposing novel approaches that are able to
accurately localize a camera are problems of high practical relevance.
This is the purpose for the proposed project, VisLocLearn.
Applicant: Torsten Sattler, Department of Electrical Engineering, Chalmers
Projects started 2019
3D Perception and Prediction of Pedestrians for Improved Descision-Taking in Autonomous Driving and active Safety
Research Project, 2018–2019
The overall purpose of the project is to basic framework for combining perception, prediction, and control to optimize system level performance. Deep learning algorithms in the vision system will be developed in connection to dynamic models of the pedestrians, where the uncertainty of the predictions will be crucial for the control performance.
PI: Fredrik Kahl, Electrical Engineering, Chalmers
Interpretable deep modeling for Systems Biology
We aim to provide AI solutions for Integromics that are effective and interpretable. To this end, an AI-Integromics approach should predict phenotypes for in-house data, attribute predictive strength to the multi-omics source data, yet also provide a mapping to extensive public data resources to facilitate drug re-purposing.
PI: Rebecka Jörnsten, Mathematical Sciences, Chalmers
Deep Reinforcement Learning: Principles and Applications in Cognitive Science and Networks
The overall project goals are to carry out basic research in Deep Reinforcemnet Learning (DRL) which is one of the most powerful and exible frameworks available today for general purpose learning, and to apply it in two areas: (a) fundamental problems in cognitive science in collaboration with the world's leading group in this area at UC Berkeley and (b) network policies for applications at Ericsson.
PI: Devdatt Dubhashi, Computer Science and Engineering, Chalmers
INNER: information theory of deep neural networks
Over the last decade, deep-learning algorithms have dramatically improved the state of the art in many machine-learning problems, including computer vision, speech recognition, natural language processing, and audio recognition. Despite their success, however, there is no satisfactory mathematical theory that explains the functioning of such algorithms. Indeed, a common critique is that deep-learning algorithms are often used as black box, which is unsatisfactory in all applications for which performance guarantees are critical (e.g., traffic-safety applications). The purpose of this project is to increase our theoretical understanding of deep neural networks. This will be done by relying on tools of information theory and focusing on specific tasks that are relevant to computer vision.
PI: Giuseppe Durisi, Electrical Engineering, Chalmers
Software Engineering for AI/ML/DL
The development and evolution of software is quite fundamentally changing. Whereas earlier software was exclusively built based on requirements, in more recent years we see that there are three distinct approaches to the creation of software, i.e. requirements-driven, outcome-driven and AI-driven.
PI: Jan Bosch, Computer Science and Engineering, Chalmers
Verification of Machine Learning Algorithms (Vermillion)
Investigate if new methods of testing of cyber-physical systems being developed in our VR Researchv Environment [SyTec,CSE18] can be adapted to work on neural networks. Study applications in ML based Natural language translation (where errors are everywhere, and grouping them is what is needed). Study testing and formal verification of ML based autonomous driving in collaboration with RiSE Viktoria. Study testing in the context of machine learning based software development (see Erik Meijer’s recent talk
at Chalmers; Meijer has agreed to be an advisor on the project). The postdoc will likely concentrate on one of these application domains, depending on their background.The long term goal is to develop methods to design neural networks and machine learning based software with guaranteed properties.
Vision and machine learning for collaborative robotics (ViMCoR)
In this project we will develop the vision capabilities of collaborative robots by using machine learning methods. In manufacturing, of cars and trucks, industrial robots have been mainly used during the body in white stages (i.e. to do welding of the main body components).
PI: Bengt Lennartson, Electrical Engineering, Chalmers