Keywords: Bayesian inference, learning and reasoning, stochastic control theory, neural networks, statistical physics,
quantum machine learning
My research focusess on the design of efficient computational methods AI and machine learning
using ideas and methods from statistical physics and quantum physics. In addition, I am interested in how intelligence and
consciousness may arise in the brain. Here, I give a high level overview of my research interests, both past and present.
For details consult my Google scholar page.
Due to the essential roles that noise and uncertainty play in perception and learning, a useful way to model intelligence
is to use probability models. In the mid 90s, the fields of analog and digital computing as separate approaches to model
intelligence, have begun to merge using the idea of Bayesian inference: One can generalize the logic of digital computation
to a probabilistic calculus, embodied in a so-called graphical model. Similarly, one can generalize dynamical systems to
stochastic dynamical systems that allow for a probabilistic description in terms of a Markov process. The Bayesian paradigm
has greatly helped to integrate different schools of thought in particular in the field of artificial intelligence and
machine learning but also provides a computational paradigm for neuroscience.
A typical Bayesian computation, whether in the context of a complex data analysis
problem or in a stochastic neural network, is to compute an expectation value, which is referred to as Bayesian inference.
Bayesian inference is intractable, which means that computation time and memory use scale exponentially with the problem
size. However, many methods exist to compute these quantities approximately. Most of these methods origin from statistical
physics, such as the mean field method, belief propagation or Monte Carlo sampling. Application of these methods to machine
learning problems is challenging and an active field of research to which I have made several contributions.
Control theory is a theory from engineering that gives a formal description of how a system, such as a robot or animal, can
move from a current state to a future state at minimal cost, where cost can mean time spent, or energy spent or any other
quantity. Control theory is used traditionally to control industrial plants, airplanes or missiles, but is also the natural
framework to model intelligent behavior in animals or robots. The mathematical formulation of deterministic control theory
is very similar to classical mechanics. In fact, classical mechanics can be viewed as a special case of control theory.
Stochastic control theory uses the language of stochastic differential equations. For a certain class of stochastic
control problems, the solution is described by a linear partial differential equation that can be solved formally as a path
integral. This so-called path integral control method provides a deep link between control, inference and statistical
physics. This statistical physics view of control theory shows that qualitative different control solutions exist for
different noise levels separated by phase transitions. The control solutions can be computed using efficient approximate
inference methods such as Monte Carlo sampling or deterministic approximation methods. The path integral control theory is
successfully being used by leading research groups in robotics world wide. For more information see the
path integral control theory page.
Current successes in machine learning has ignited interesting new
connections between machine learning and quantum physics, loosely referred to
as quantum machine learning.
learning methods are finding useful applications in quantum physics, such as
characterizing the ground state of a quantum Hamiltonian or to learn
different phases of matter.
Since 2018, I am interested in how the quantum formalism can be used to advance
machine learning. Recent work:
Quantum Boltzmann Machine
One line of work is the quantum Boltzmann machine, which is a method to learn a quantum Hamiltonian from classical or
quantum data. The learning rule requires the computation of quantum spin expectation values which is intractable, as in the
In this paper we propose a new method to accellerate learning using a quantum circuit
Adiabatic quantum computing
Another line of work is to explore the possibility of quantum advantage using adiabatic quantum computing. In particular,
we generalize the well-known quadratic speedup of Grover search to general optimization problems.
We show that this is in principle possible, but that in practice this faces two serious obstacles. The speedup is
achievable using an optimized annealing schedule that requires the exact value of annealing parameter at the phase
transition. Computation of this number is intractable in general. Secondly, the value needs to be specified with with a
numerical precision that increases exponentially with the problem size
Since 2020 we have an intense new collaboration with the
scanning tunneling microscopy group of professor Alex
Khajetoorians. In this collaboration, we have shown the
possibility to realize a stochastic neural network at atomic scale.
The spins in this network are bi-stable atoms that stochastically switch
between two states (up and down). Each spin or neuron is characterized by the
asymmetry (the probability to be in the up state minus down state) and
mean residence time (the mean time between switches). Residence times
of different spins can differ many orders of magnitude. We
proposed that fast spins encode the firing or non-firing of neurons and
slow spins encode binary learning elements, ie. synapses. In this way, a physical substrate
can implement learning as the long term change of the slow variables.
The idea to assist medical doctors to diagonse patients based on their symptoms is one of the
oldest ideas of the use of artificial intelligence. However, up to today, building such systems with high accuracy has
proven surprisingly difficult. In collaboration with the Erasmus MC in Rotterdam, we are building such an expert system for
the diagnosis of internal medicine related diseases as they occur in the emergency department. The system is based on a
Bayesian network that is specified on the basis of the knowledge of medical experts and textbooks.
There is an exciting possibility to use a quantum mechanical wave function to represent a
probability distribution. While classically the probability distribution p(x) is computed for
each x separately, the quantum physics computes all 'compoonents' of the wave function
simulatenously and in parallel. This implies that the computation of statistics (means,
correlations) of high dimensional distribution, which requires exponentially long computation
times using classical machines, could be computed in constant time on a quantum device.
My recent work focusses on learning such quantum systems. The learning step requires the
estimation of the above statistics and is done classically using Monte Carlo sampling. The
long term aim is to replace this step by a quantum computer.
The use of the quantum formalism for learning also yields novel quantum statistics for purely
classical data analysis. These statistics signal entanglement in classical data. The research
focuses on 1) developing fast approximate inference methods for quantum learning 2) data
analysis using quantum statistics.
Sparse regression with the Garrote
Standard learning problems are to explain a dependent variable ('the output') in terms of independent variables ('the inputs').
In many learning problems, the number of input variables is large compared to the number of available data samples. Examples are found in genetics, neuro imaging and in general in many pattern recognition problems.
In order to obtain a reasonable solution in these cases, the problem needs to be regularized, typically by adding a constraint that enforces a solution with small norm.
In addition often a sparse solution is desired, which explains the output in terms of a (small) subset of the input variables.
The Lasso method is a sparse regression method that uses an L1 norm as regularizer. The Lasso is very fast and can be applied to very large problems. However, the method suffers from
'shrinkage' which means that in certain cases the wrong inputs are identified. Ideally, one would use a regularizes which penalizes the number of inputs rather than their strength. This is achieved using the so-called L0 regularizer. However, to find the solution in this case is significantly more difficult. Examples of approaches are Monte Carlo methods or the
All sparse methods suffer from strongly correlated inputs. Examples are the spatial correlations between nearby genetic measurements, or pixels in images.
In this project, the student will extend the variational garrote to take these correlations into account and to demonstrate the improved performance on neuro-imaging or genetic data.
Data analyis for sustainable energy consumption
In collaboration with NRLytics, a young start-up in the energy sector, this project aims to
use machine learning methods to analyse and optimize energy consumption. See Project description (in Dutch)
Kees Albers was PhD student and postdoc on approximate
inference methods for genetic linkage analysis. Kees was 4 years at Sanger
Institute, Cambridge UK and is since 2012 at Human Genetics in Nijmegen.
Bram Kasteel was Bachelor student on the topic of multi-agent control
Stijn Tonk was Master student on the topic of multi-agent control
Ender Akay was programmer for Smart Research bv and Promedas bv
Gulliver de Boer was Bachelor student on the topic of multi-agent control applied to poker
Max Bakker was a Bachelor student on the topic of multi-agent systems
Ben Ruijl was a Bachelor student on the topic of multi-agent systems
Henk Griffioen was Master student on the topic of genetic association studies
Bart van den broek was PhD student on the topic of stochastic optimal control theory
Takamitsu Matsubara is assistant professor of robotics at the Nara Institue of Science
and Technology on sabatical leave in our group in 2013
Alberto Llera was PhD student on the topic of Brain Computer Interfaces, now postdoc with Christian Beckman at the Donders Center for Imaging
Satoshi Satoh is assistant professor at the
faculty of engineering of
Hiroshima University in Japan. He visited in 2011-2012 to generalize the path integral control
method and to apply this method to concrete problems in control and robotics.
Sep Thijssen was a PhD student funded by Thales Nederland and on the Complacs project, working on application of
stochastic optimal control methods for multi-agent systems. See here his very readable PhD Thesis.
Han Nauta was Master student working on path integral control problems
Hans Ruiz was a PhD student on the NETT project. He works on the application of stochastic optimal control methods in
neuroscience for multi-agent systems
Thalmeier was a PhD student on the NETT project. He works on the application of stochastic optimal control methods in
neuroscience for multi-agent systems
Silvia Menchon is assistant professor at the University of Cordoba (Argentina), visiting in 2015-2016 funded by the Radboud Excellence Initiative.