Publications

Abstract taken from Google Scholar:

Advances in reinforcement learning (RL) often rely on massive compute resources and remain notoriously sample inefficient. In contrast, the human brain is able to efficiently learn effective control strategies using limited resources. This raises the question whether insights from neuroscience can be used to improve current RL methods. Predictive processing is a popular theoretical framework which maintains that the human brain is actively seeking to minimize surprise. We show that recurrent neural networks which predict their own sensory states can be leveraged to minimise surprise, yielding substantial gains in cumulative reward. Specifically, we present the Predictive Processing Proximal Policy Optimization (P4O) agent; an actor-critic reinforcement learning agent that applies predictive processing to a recurrent variant of the PPO algorithm by integrating a world model in its hidden state. P4O significantly outperforms a baseline recurrent variant of the PPO algorithm on multiple Atari games using a single GPU. It also outperforms other state-of-the-art agents given the same wall-clock time and exceeds human gamer performance on multiple games including Seaquest, which is a particularly challenging environment in the Atari domain. Altogether, our work underscores how insights from the field of neuroscience may support the development of more capable and efficient artificial agents.

Go to article

Abstract taken from Google Scholar:

Visual neuroprostheses are a promising approach to restore basic sight in visually impaired people. A major challenge is to condense the sensory information contained in a complex environment into meaningful stimulation patterns at low spatial and temporal resolution. Previous approaches considered task-agnostic feature extractors such as edge detectors or semantic segmentation, which are likely suboptimal for specific tasks in complex dynamic environments. As an alternative approach, we propose to optimize stimulation patterns by end-to-end training of a feature extractor using deep reinforcement learning agents in virtual environments. We present a task-oriented evaluation framework to compare different stimulus generation mechanisms, such as static edge-based and adaptive end-to-end approaches like the one introduced here. Our experiments in Atari games show that stimulation patterns obtained via task-dependent end-to-end optimized reinforcement learning result in equivalent or improved performance compared to fixed feature extractors on high difficulty levels. These findings signify the relevance of adaptive reinforcement learning for neuroprosthetic vision in complex environments.

Go to article

Abstract taken from Google Scholar:

Neural prosthetics may provide a promising solution to restore visual perception in some forms of blindness. The restored prosthetic percept is rudimentary compared to normal vision and can be optimized with a variety of image preprocessing techniques to maximize relevant information transfer. Extracting the most useful features from a visual scene is a nontrivial task and optimal preprocessing choices strongly depend on the context. Despite rapid advancements in deep learning, research currently faces a difficult challenge in finding a general and automated preprocessing strategy that can be tailored to specific tasks or user requirements. In this paper, we present a novel deep learning approach that explicitly addresses this issue by optimizing the entire process of phosphene generation in an end-to-end fashion. The proposed model is based on a deep auto-encoder architecture and includes a highly adjustable simulation module of prosthetic vision. In computational validation experiments, we show that such an approach is able to automatically find a task-specific stimulation protocol. The results of these proof-of-principle experiments illustrate the potential of end-to-end optimization for prosthetic vision. The presented approach is highly modular and our approach could be extended to automated dynamic optimization of prosthetic vision for everyday tasks, given any specific constraints, accommodating individual requirements of the end-user.

Go to article

Abstract taken from Google Scholar:

Neuroprosthetic implants are a promising technology for restoring some form of vision in people with visual impairments via electrical neurostimulation in the visual pathway. Although an artificially generated prosthetic percept is relatively limited compared with normal vision, it may provide some elementary perception of the surroundings, re-enabling daily living functionality. For mobility in particular, various studies have investigated the benefits of visual neuroprosthetics in a simulated prosthetic vision paradigm with varying outcomes. The previous literature suggests that scene simplification via image processing, and particularly contour extraction, may potentially improve the mobility performance in a virtual environment. In the current simulation study with sighted participants, we explore both the theoretically attainable benefits of strict scene simplification in an indoor environment by controlling the environmental complexity, as well as the practically achieved improvement with a deep learning-based surface boundary detection implementation compared with traditional edge detection. A simulated electrode resolution of 26× 26 was found to provide sufficient information for mobility in a simple environment. Our results suggest that, for a lower number of implanted electrodes, the removal of background textures and within-surface gradients may be beneficial in theory. However, the deep learning-based implementation for surface boundary detection did not improve mobility performance in the current study. Furthermore, our findings indicate that, for a greater number of electrodes, the removal of within-surface gradients and background textures …

Go to article

Abstract taken from Google Scholar:

Visual neuroprostheses are a promising approach to restore basic sight in visually impaired people. A major challenge is to condense the sensory information contained in a complex environment into meaningful stimulation patterns at low spatial and temporal resolution. Previous approaches considered task-agnostic feature extractors such as edge detectors or semantic segmentation, which are likely suboptimal for specific tasks in complex dynamic environments. As an alternative approach, we propose to optimize stimulation patterns by end-to-end training of a feature extractor using deep reinforcement learning agents in virtual environments. We present a task-oriented evaluation framework to compare different stimulus generation mechanisms, such as static edge-based and adaptive end-to-end approaches like the one introduced here. Our experiments in Atari games show that stimulation patterns obtained via task-dependent end-to-end optimized reinforcement learning result in equivalent or improved performance compared to fixed feature extractors on high difficulty levels. These findings signify the relevance of adaptive reinforcement learning for neuroprosthetic vision in complex environments.

Go to article

Abstract taken from Google Scholar:

Although the introduction of exome sequencing (ES) has led to the diagnosis of a significant portion of patients with neurodevelopmental disorders (NDDs), the diagnostic yield in actual clinical practice has remained stable at approximately 30%. We hypothesized that improving the selection of patients to test on the basis of their phenotypic presentation will increase diagnostic yield and therefore reduce unnecessary genetic testing.We tested 4 machine learning methods and developed PredWES from these: a statistical model predicting the probability of a positive ES result solely on the basis of the phenotype of the patient.We first trained the tool on 1663 patients with NDDs and subsequently showed that diagnostic ES on the top 10% of patients with the highest probability of a positive ES result would provide a diagnostic yield of 56%, leading to a notable 114% increase. Inspection of our …

Go to article

Abstract taken from Google Scholar:

While both molecular and phenotypic data are essential when interpreting genetic variants, prediction scores (CADD, PolyPhen, and SIFT) have focused on molecular details to evaluate pathogenicity - omitting phenotypic features. To unlock the full potential of phenotypic data, we developed PhenoScore: an open source, artificial intelligence-based phenomics framework. PhenoScore combines facial recognition technology with Human Phenotype Ontology (HPO) data analysis to quantify phenotypic similarity at both the level of individual patients as well as of cohorts. We prove PhenoScore9s ability to recognize distinct phenotypic entities by establishing recognizable phenotypes for 25 out of 26 investigated genetic syndromes against clinical features observed in individuals with other neurodevelopmental disorders. Moreover, PhenoScore was able to provide objective clinical evidence for two distinct ADNP-related phenotypes, that had already been established functionally, but not yet phenotypically. Hence, PhenoScore will not only be of use to unbiasedly quantify phenotypes to assist genomic variant interpretation at the individual level, such as for reclassifying variants of unknown clinical significance, but is also of importance for detailed genotype-phenotype studies.

Go to article

Abstract taken from Google Scholar:

Background and Objective: Since several genetic disorders exhibit facial characteristics, facial recognition techniques can help clinicians in diagnosing patients. However, currently, there are no open-source models that are feasible for use in clinical practice, which makes clinical application of these methods dependent on proprietary software. Methods: In this study, we therefore set out to compare three facial feature extraction methods when classifying 524 individuals with 18 different genetic disorders: two techniques based on convolutional neural networks (VGGFace2, OpenFace) and one method based on facial distances, calculated after detecting 468 landmarks. For every individual, all three methods are used to generate a feature vector of a facial image. These feature vectors are used as input to a Bayesian softmax classifier, to see which feature extraction method would generate the best results. Results: Of the considered algorithms, VGGFace2 results in the best performance, as shown by its accuracy of 0.78 and significantly lowest loss. We inspect the features learned by VGGFace2 by generating activation maps and using Local Interpretable Model-agnostic Explanations, and confirm that the resulting predictors are interpretable and meaningful. Conclusions: All in all, the classifier using the features extracted by VGGFace2 shows not only superior classification performance, but detects faces in almost all images that are processed, in seconds. By not retraining VGGFace2, but instead using the feature vector of the network with its pretrained weights, we avoid overfitting the model. We confirm that it is possible to classify individuals with …

Go to article

Abstract taken from Google Scholar:

Speech decoding from brain activity can enable development of brain-computer interfaces (BCIs) to restore naturalistic communication in paralyzed patients. Previous work has focused on development of decoding models from isolated speech data with a clean background and multiple repetitions of the material. In this study, we describe a novel approach to speech decoding that relies on a generative adversarial neural network (GAN) to reconstruct speech from brain data recorded during a naturalistic speech listening task (watching a movie). We compared the GAN-based approach, where reconstruction was done from the compressed latent representation of sound decoded from the brain, with several baseline models that reconstructed sound spectrogram directly. We show that the novel approach provides more accurate reconstructions compared to the baselines. These results underscore the potential of GAN …

Go to article

Abstract taken from Google Scholar:

Quasi-experimental research designs, such as regression discontinuity and interrupted time series, allow for causal inference in the absence of a randomized controlled trial, at the cost of additional assumptions. In this paper, we provide a framework for discontinuity-based designs using Bayesian model averaging and Gaussian process regression, which we refer to as ‘Bayesian nonparametric discontinuity design’, or BNDD for short. BNDD addresses the two major shortcomings in most implementations of such designs: overconfidence due to implicit conditioning on the alleged effect, and model misspecification due to reliance on overly simplistic regression models. With the appropriate Gaussian process covariance function, our approach can detect discontinuities of any order, and in spectral features. We demonstrate the usage of BNDD in simulations, and apply the framework to determine the effect of running for political positions on longevity, of the effect of an alleged historical phantom border in the Netherlands on Dutch voting behaviour, and of Kundalini Yoga meditation on heart rate.

Go to article

Abstract taken from Google Scholar:

Deep neural networks (DNNs) are an indispensable machine learning tool despite the difficulty of diagnosing what aspects of a model’s input drive its decisions. In countless real-world domains, from legislation and law enforcement to healthcare, such diagnosis is essential to ensure that DNN decisions are driven by aspects appropriate in the context of its use. The development of methods and studies enabling the explanation of a DNN’s decisions has thus blossomed into an active and broad area of research. The field’s complexity is exacerbated by competing definitions of what it means “to explain” the actions of a DNN and to evaluate an approach’s “ability to explain”. This article offers a field guide to explore the space of explainable deep learning for those in the AI/ML field who are uninitiated. The field guide: i) Introduces three simple dimensions defining the space of foundational methods that contribute to explainable deep learning, ii) discusses the evaluations for model explanations, iii) places explainability in the context of other related deep learning research areas, and iv) discusses user-oriented explanation design and future directions. We hope the guide is seen as a starting point for those embarking on this research field.

Go to article

Abstract taken from Google Scholar:

Neural prosthetics may provide a promising solution to restore visual perception in some forms of blindness. The restored prosthetic percept is rudimentary compared to normal vision and can be optimized with a variety of image preprocessing techniques to maximize relevant information transfer. Extracting the most useful features from a visual scene is a nontrivial task and optimal preprocessing choices strongly depend on the context. Despite rapid advancements in deep learning, research currently faces a difficult challenge in finding a general and automated preprocessing strategy that can be tailored to specific tasks or user requirements. In this paper, we present a novel deep learning approach that explicitly addresses this issue by optimizing the entire process of phosphene generation in an end-to-end fashion. The proposed model is based on a deep auto-encoder architecture and includes a highly adjustable simulation module of prosthetic vision. In computational validation experiments, we show that such an approach is able to automatically find a task-specific stimulation protocol. The results of these proof-of-principle experiments illustrate the potential of end-to-end optimization for prosthetic vision. The presented approach is highly modular and our approach could be extended to automated dynamic optimization of prosthetic vision for everyday tasks, given any specific constraints, accommodating individual requirements of the end-user.

Go to article

Abstract taken from Google Scholar:

A fundamental aspect of human experience is that it is segmented into discrete events. This may be underpinned by transitions between distinct neural states. Using an innovative data-driven state segmentation method, we investigate how neural states are organized across the cortical hierarchy and where in cortex neural state and perceived event boundaries overlap. Our results show that neural state boundaries are organized in a temporal cortical hierarchy, with short states in primary sensory regions and long states in anterior temporal pole and lateral and medial prefrontal cortex. Neural state boundaries overlap with event boundaries across large parts of this hierarchy. State boundaries are shared within and between groups of brain regions that resemble well known functional networks, such as the default mode network that fractionates into two subnetworks–one fast, one slow. Together these findings suggest that a nested cortical hierarchy of neural states forms the basis of event segmentation.

Go to article

Abstract taken from Google Scholar:

Neuroprosthetic implants are a promising technology for restoring some form of vision in people with visual impairments via electrical neurostimulation in the visual pathway. Although an artificially generated prosthetic percept is relatively limited compared with normal vision, it may provide some elementary perception of the surroundings, re-enabling daily living functionality. For mobility in particular, various studies have investigated the benefits of visual neuroprosthetics in a simulated prosthetic vision paradigm with varying outcomes. The previous literature suggests that scene simplification via image processing, and particularly contour extraction, may potentially improve the mobility performance in a virtual environment. In the current simulation study with sighted participants, we explore both the theoretically attainable benefits of strict scene simplification in an indoor environment by controlling the environmental complexity, as well as the practically achieved improvement with a deep learning-based surface boundary detection implementation compared with traditional edge detection. A simulated electrode resolution of 26× 26 was found to provide sufficient information for mobility in a simple environment. Our results suggest that, for a lower number of implanted electrodes, the removal of background textures and within-surface gradients may be beneficial in theory. However, the deep learning-based implementation for surface boundary detection did not improve mobility performance in the current study. Furthermore, our findings indicate that, for a greater number of electrodes, the removal of within-surface gradients and background textures …

Go to article

Abstract taken from Google Scholar:

Neural decoding can be conceptualized as the problem of mapping brain responses back to sensory stimuli via a feature space. We introduce (i) a novel experimental paradigm that uses well-controlled yet highly naturalistic stimuli with a priori known feature representations and (ii) an implementation thereof for HYPerrealistic reconstruction of PERception (HYPER) of faces from brain recordings. To this end, we embrace the use of generative adversarial networks (GANs) at the earliest step of our neural decoding pipeline by acquiring fMRI data as participants perceive face images synthesized by the generator network of a GAN. We show that the latent vectors used for generation effectively capture the same defining stimulus properties as the fMRI measurements. As such, these latents (conditioned on the GAN) are used as the in-between feature representations underlying the perceived images that can be …

Go to article