jeff garzik bitcoin stock price
The availability of an open-ended vista of admissible ways to achieve one's goals constitutes a good operational definition of "awareness" of those goals. What researchers mean by this is that enhancements might be not only to the database of things a machine can do, but to its algorithms for deciding minerva to do. But once computer technology is out there, quantum will get ever bitcoin and filter down to hobbyists, hackers, and "machine rights" organizations. For example, knowledge may be factual or propositional: Nor would they be constrained to organize their society, and its rules, as do we.
is bitcoin mining profitable 2014 dodge ram 1500
Thought experiments about these matters are the source of practical insights into human and machine behavior and suggest how to build different and better kinds of machines. If Big Blue beat Kasparov when he was one of the strongest world champion chess players ever, he and most observers believe that even better chess is played by teams of humans and machines combined. Can natural and artificial selection be programmed into self-replicating robots? Students will need to have a working understanding of penetration testing techniques. The student may choose the algorithms that underly the engine, such as alpha-beta search, Monte-Carlo tree search, or neural networks.
kim kardashian pumpkin carving В»
cara cepat dapat bitcoin kaskus
But science is a long way from unlocking the secrets in nature's infinite book. This can apply to both lossless and lossy compression. University of Oxford MSc Dissertation, This project bitcoin investigate a rich research line, recently pursued by a few within the Department of CS, looking at the development of minerva abstractions of Markovian models. This project is computer for someone with at quantum basic knowledge of machine learning.
jaxx bitcoin and ethereum wallet В»
Today's Stock Market News and Analysis - jcadesigns.gogarraty.com
GeomLab has a turtle graphics feature, but the pictures are drawn only on the screen. It should be possible to make a turtle out of Lego Mindstorms, then control it with an instance of Geomlab running on a host computer, with communication over Bluetooth.
Either use an interpreter for GeomLab's intermediate code to execute GeomLab programs, or investigate dynamic translation of the intermediate code into code for Android's virtual machine Dalvik.
At present, Keiko supports only conventional Pascal-like language implementations that store activation records on a stack. Experiment with an implementation where activation records are heap-allocated and therefore recovered by a garbage collector , procedures are genuinely first-class citizens that can be returned as results in addition to being passed as arguments, and tail recursion is optimised seamlessly.
Alternative firmware for the Mindstorms robot controller provides an implementation of the JVM, allowing Java programs to run on the controller, subject to some restrictions.
Using this firmware as a guide, produce an interpreter for a suitable bytecode, perhaps some variant of Keiko, allowing Oberon or another robot language of your own design to run on the controller.
Aim to support the buttons and display at first, and perhaps add control of the motors and sensors later. The GeomLab language is untyped, leading to errors when expressions are evaluated that would be better caught at an earlier stage.
Most GeomLab programs, however, follow relatively simple typing rules. The aim in this project is to write a polymorphic type checker for GeomLab and integrate it into the GeomLab system, which is implemented in Java. A simple implementation of the type-checker would wait until an expression is about to be evaluated, and type-check the whole program at that point.
As an extension of the project, you could investigate whether it is possible to type-check function definitions one at a time, even when some of the functions they call have not yet been defined. The guarded negation fragment of first-order logic is an expressive logic of interest in databases and knowledge representation. It has been shown to have a decidable satisfiability problem but, to the best of my knowledge, there is no tool actually implementing a decision procedure for it.
The goal would be to design a tool to determine whether or not a formula in this logic is satisfiable. Most likely, this would require designing and implementing a tableau-based algorithm, in the spirit of related tools for description logics and the guarded fragment. Logic and Proof or equivalent. There are some connections to material in Knowledge Representation and Reasoning, but this is not essential background. Let F1 and F2 be sentences in first-order logic, say such that F1 entails F2: The goal in this project is to explore and implement procedures for constructing interpolants, particularly for certain decidable fragments of first-order logic.
It turns out that finding interpolants like this has applications in some database query rewriting problems. Computed tomography CT scanning is a ubiquitous scanning modality. It produces volumes of data representing internal parts of a human body. The slices most frequently come at a resolution of x voxels, achieving an accuracy of about 0. The distance between slices is a parameter of the scanning process and is typically much larger, about 5mm.
During the analysis of CT data volumes it is often useful to correct for the large spacing between slices.
For example when preparing a model for 3D printing, the axial voxels would appear elongated. These could be corrected through an interpolation process along the spinal axis. This project is about the interpolation process, either in the raw data output by the scanner, or in the post-processed data which is being prepared for further analysis or 3D printing.
The output models would ideally be files in a format compatible with 3D printing, such as STL. The main aesthetic feature of the output would be measurable as a smoothness factor, parameterisable by the user.
Isolating the complex roots of a polynomial can be achieved using subdivision algorithms. Traditional Newton methods can be applied in conjunction with interval arithmetic.
Moore's, Krawczyk's and Hansen-Sengupta's. CORE defines multiple levels of operation over which a program can be compiled and executed. Each of these levels provide stronger guarantees on exactness, traded against efficiency. Further extensions of this work can include and are not limited to: The code has been included and is available within the CORE repository.
Scientists in the Experimental Psychology Department study patients with a variety of motor difficulties, including apraxia - a condition usually following stroke which involves lack of control of a patient over their hands or fingers.
Diagnosis and rehabilitation are traditionally carried out by Occupational Therapists. In recent years, computer-based tests have been developed in order to remove the human subjectivity from the diagnosis, and in order to enable the patient to carry out a rehabilitation programme at home. One such test involves users being asked to carry out static gestures above a Leap Motion sensor, and these gestures being scored according to a variety of criteria.
A prototype has been constructed to gather data, and some data has been gathered from a few controls and patients. In order to deploy this as a clinical tool into the NHS, there is need for a systematic data collection and analysis tool, based on machine learning algorithms to help classify the data into different categories. Algorithms are also needed in order to classify data from stroke patients, and to assess the degree of severity of their apraxia. Also, the graphical user interface needs to be extended to give particular kinds of feedback to the patient in the form of home exercises, as part of a rehabilitation programme.
Due to Glyn's untimely death a new co-supervisor needs to be found in the Experimental Psychology Department. It is unrealistic to assume this project can run in the summer of Psychology has inspired and informed a number of machine learning methods. Decisions within an algorithm can be made so as to improve an overall aim of maximising a cumulative reward.
Supervised learning methods in this class are known as Reinforcement Learning. A basic reinforcement learning model consists of establishing a number of environment states, a set of valid actions, and rules for transitioning between states. Applying this model to the rules of a board game means that the machine can be made to learn how to play a simple board game by playing a large number of games against itself.
The goal of this project is to set up a reinforcement learning environment for a simple board game with a discrete set of states such as Backgammon. If time permits, this will be extended to a simple geometric game such as Pong where the states may have to be parameterised in terms of geometric actions to be taken at each stage in the game.
One such test involves users drawing simple figures on a tablet, and these figures being scored according to a variety of criteria. Data has already been gathered from or so controls, and is being analysed for a range of parameters in order to assess what a neurotypical person could achieve when drawing such simple figures.
Further machine learning analysis could help classify such data into different categories. Knee replacement surgery involves a precise series of steps that a surgeon needs to follow. Trainee surgeons have traditionally mastered these steps by learning from textbooks or experienced colleagues.
It is proposed to construct a computer-based tool which would help with this goal. Apart from the choice of tools and materials, the tool would also feature a virtual model of the knee.
The graphical user interface will present a 3D model of a generic knee to be operated, and would have the ability for the user to make cuts necessary to the knee replacement procedure. There would be pre-defined parameters regarding the type and depth of each cut, and an evaluation tool on how the virtual cuts compared against the parameters. The project goals are quite extensive and so this would be suitable for an experienced programmer.
The goal of this project is to write a program that model checks a Markov chain against an LTL formula, i.
The two main algorithmic tasks are to efficiently compile LTL formulas into automata and then to solve systems of linear equations arising from the product of the Markov chain and the automaton.
An important aspect of this project is to make use of an approach that avoids determinising the automaton that represents the LTL formula. An optimal automata approach to LTL model checking of probabilistic systems. The interval program analysis is a well-known algorithm for estimating the behaviour of programs without actually running them.
This algorithm assumes that all the functions called by the input program are defined in the program, so that the source code of every called function can be found in the program. However, in practice, this assumption is not necessarily met. Programs often use library functions whose source code is not available. The goal of this project is to lift this assumption.
During the project, a student will develop an interval-analysis algorithm that works in the presence of calls to unknown library functions, implement the algorithm, and evaluate the algorithm experimentally. Undergraduate students who wish to enquire about a project for are welcome to contact Prof Yang but should note that the response may be delayed as he is on sabbatical.
Recently, researchers in machine learning have developed new Turing-complete languages, such as Infer. The goal of this project is to study these languages using tools from programming languages.
Specifically, a student will work on developing a new inference algorithm for probabilistic programs that mix techniques from program analysis and those from the Monte Carlo simulation, a common method for performing inference on probabilistic programs.
Or the student will explore the connection between the use of computational effects in higher-order functional probabilistic programming languages and the encoding of advanced probability models in those languages in particular, nonparametric Bayesian models , which has been pointed out by the recent work of Dan Roy and his colleagues.
Compiler and Machine learning courses. The Programming Language course is not required, but useful for carrying out this project. This can further lead to optimised maintenance for the building devices.
Of course the sensitisation of buildings leads to heavy requirements on the overall infrastructure: Further, we plan to investigate approaches to perform meta-sensing, namely to extrapolate the knowledge from physical sensors towards that of virtual elements as an example, to infer the current building occupancy from correlated measurements of temperature and humidity dynamics. On the actuation side, we are likewise interested in engineering non-invasive minimalistic solutions, which are robust to uncertainty and performance-certified.
The plan for this project is to make the first steps in this direction, based on recent results in the literature. The project can benefit from a visit to Honeywell Labs Prague.
Some familiarity with dynamical systems. The idea is that we can explore a pair of structures, e. If we can always keep these pebbles in sync so that the two k-sized windows look the same are isomorphic then we say that Duplicator has a winning strategy for the k-pebble game. This gives a resource-bounded notion of approximation to graphs and other structures which has a wide range of applications. Monads and comonads are widely used in functional programming, e.
It turns out that pebble games, and similar notions of approximate or local views on data, can be captured elegantly by comonads, and this gives a powerful language for many central notions in constraints, databases and descriptive complexity. Finally, monads can be used to give various notions of approximate or non-classical solutions to computational problems.
These include probabilistic and quantum solutions. The aim of this project is to explore a number of aspects of these ideas. Depending on the interests and background of the student, different aspects may be emphasised, from functional programming, category theory, logic, algorithms and descriptive complexity, probabilistic and quantum computation. Some specific directions include: Developing Haskell code for the k-pebbling comonad and various non-classical monads, and using this to give a computational tool-box for various constructions in finite model theory and probabilistic or quantum computation.
Developing the category-theoretic ideas involved in combining monads and comonads, and studying some examples. Using the language of comonads to capture other important combinatorial invariants such as tree-depth. Developing the connections between category theory, finite model theory and descriptive complexity.
Leonid Libkin, Elements of finite model theory. Background on pebble games and the connection with logic and complexity. The pebbling comonad in finite mode theory. Technical report describing the basic ideas which can serve as a starting point. Using dynamic traffic analysis techniques, we have mapped many of the third-party data flows, including particular data types, from popular applications. The aim of this project would be to extend this work by deploying a mobile application that automatically reveals to the user what kinds of data are sent to whom via the apps they installed on their device.
Alternatively, the project could look into extending our existing traffic analysis framework in order to scale up the analysis and improve the accuracy and coverage of the tracker detection. The goal of this project is to take an existing scheduling program and a class of real-life industrial problems and to develop a visualisation program that could help an end-user picture the running of a particular schedule as a three-dimensional animation. The skill-set required of a student taking this project would primarily be in three-dimensional animation, either using their own code or by bolting onto an existing animation tool such as Blender.
It should also be possible for the end-user to easily vary the context of the visualisation i. An on-line example of such a visualisation is at https: Manual manipulation of such graphs is slow and error prone.
This project employs a formalism, based on monoidal categories, that supports mechanised reasoning with open-graphs. This gives a compositional account of graph rewriting that preserves the underlying categorical semantics. Concurrency, Concurrent Programming, Computer Security all possibly an advantage. In practice, datasources often contain sensitive information that the data owners want to keep inaccessible to users.
In a recent research paper, the project supervisors have formalized and studied the problem of determining whether a given data integration system discloses sensitive information to an attacker. The paper studies the computational properties of the relevant problems and also identifies situations in which practical implementations are feasible.
The goal of the project is to design and implement practical algorithms for checking whether information disclosure can occur in a data integration setting. These algorithms would be applicable to the aforementioned situations for which practical implementations seem feasible. There are several possible locations, and each person prefers to have the facility as close to them as possible.
However, the central planner, who will make the final decision, does not know the voters' location, and, moreover, for various reasons such as privacy or the design of the voting machine , the voters cannot communicate their location. Instead, they communicate their ranking over the available locations, ranking them from the closest one to the one that is furthest away. The central planner then applies a fixed voting rule to these rankings.
The quality of each location is determined by the sum of distances from it to all voters, or alternatively the maximum distance. A research question that has recently received a substantial amount of attention is whether classic voting rules tend to produce good-quality solutions in this setting.
The goal of the project would be to empirically evaluate various voting rules with respect to this measure, both for single-winner rules and for multi-winner rules where more than one facility can be opened. In addition to purely empirical work, there are interesting theoretical questions that one could explore, such as proving worst-case upper and lower bounds of the performance of various rules.
With all their feature richness, they enrich our personal online experience and provide some great new challenges for research. Based on a number of production sites from one or two domains, we will build our corpus of web interfaces, connected to a shared database.
OPAL determines the meaning of individual form elements, e. Over the course of this MSC project, we will develop a tool which invokes OPAL to analyze a given form, to explore all available submission mechanisms on this form, analyze the resulting pages for forms continuing the initial query, and to combine the outcome all found forms into a single interaction description. Benchmarks for Bayesian Deep Learning: A main challenge in BDL is comparing different tools to each other, with common benchmarks being much needed in the field.
In this project we will develop a set of tools to evaluate Bayesian deep learning techniques, reproduce common techniques in BDL, and evaluate them with the developed tools.
The tools we will develop will rely on downstream tasks that have made use of BDL in real-world applications such as parameter estimation in Strong Gravitational Lensing with neural networks. The tools we will develop will rely on downstream tasks that have made use of BDL in real-world applications such as detecting diabetic retinopathy from fundus photos and referring the most uncertain decisions for further inspection.
The requirement for large amounts of data forms a major hurdle in using RL algorithms for tasks in robotics though, where each real-world experiment would cost time and potential damage to the robot.
In this project we will develop a mock "Challenge" similar to Kaggle challenges. In this challenge we will restrict the amount of data a user can query the system at each point in time, and try to implement simple RL baselines under this constraint.
We will inspect the challenge definition and try to improve it. Typically the system under study will either be undergoing time varying changes which can be recorded, or the system will have a time varying signal as input and the response signal will be recorded. Familiar everyday examples of the former include ECG and EEG measurements which record the electrical activity in the heart or brain as a function of time , whilst examples of the latter occur across scientific research from cardiac cell modelling to battery testing.
Such recordings contain valuable information about the underlying system under study, and gaining insight into the behaviour of that system typically involves building a mathematical or computational model of that system which will have embedded within in key parameters governing system behaviour. The problem that we are interested in is inferring the values of these key parameter through applications of techniques from machine learning and data science.
Application domains of current interest include modelling of the cardiac cell for assessing the toxicity of new drugs , understanding how biological enzymes work for application in developing novel fuel cells , as well as a range of basic science problems. Prerequisites familiarity with basic probability theory. Interest in working with probability is important.
We lack a model which considers the different contexts that exist in current systems, which would underpin a measurement system for determining the level of privacy risk that might be faced. This project would seek to develop a prototype model — based on a survey of known privacy breaches and common practices in data sharing.
The objective being to propose a method by which privacy risk might be considered taking into consideration the variety of threat and data-sharing contexts that any particular person or organisation might be subjected to. It is likely that a consideration of the differences and similarities of the individual or organisational points of view will need to be made, since the nature of contexts faced could be quite diverse. What is its nature? What harms might be effected? Can we build data-analytics that are resistant to such attacks, can we detect them?
It is highly unlikely that techniques for handling erroneous data will be sufficient since we are likely to face highly targeted data-corruption. Conducted with a view to exploring the minimal sets that would result in the threat detection, and producing guidance that is aimed at determining the critical datasets required for the control to be effective.
In principle, this allows viewers to obtain more accurate representations of real-world environments and objects. Naturally, HDRI would be of interest to museum curators to document their objects, particularly, non-opaque objects or whose appearance significantly alter dependent on amount of lighting in the environment.
Currently, few tools exist that aid curators, archaeologists and art historians to study objects under user-defined parameters to study those object surfaces in meaningful ways.
In this project the student is free to approach the challenge as they see fit, but would be expected to design, implement and assess any tools and techniques they develop. The student will then develop techniques to study these objects under user-specified conditions to enable curators and researchers study the surfaces of these objects in novel ways.
These methods may involve tone mapping or other modifications of light exponents to view objects under non-natural viewing conditions to have surface details stand out in ways that are meaningful to curators. In this project the student will develop an application to use such peripherals. The student is free to approach the challenge as they see fit, but would be expected to design, implement and assess the tools they develop.
The tool would also need to serve a meaningful purpose. These projects tend to have a strong focus on human-computer interaction elements, particularly on designing and implementing user-friendly and meaningful motion gestures for a variety of real-world applications. Past students for instance have developed support for Myo on Android devices so one does not have to touch a tablet screen while cooking as well as added leap motion and Kinect support to cyber security visualization tools.
Other projects on novel human-computer interfaces possible, depending on interest and inspiration. There is scope to study in more detail the global trends in capacity building in cybersecurity, the nature of the work and the partnerships that exist to support it.
An interesting analysis might be to identify what is missing through comparison with the Cybersecurity Capacity Maturity Model, a key output of the Centre , and also to consider how strategic, or not, such activities appear to be.
An extension of this project, or indeed a second parallel project, might seek to perform a comparison of the existing efforts with the economic and technology metrics that exist for countries around the world, exploring if the data shows any relationships exist between those metrics and the capacity building activities underway.
This analysis would involve regression techniques. The result being a report highlighting various exploitable weak-points and how they might result in unauthorised access should a malign entity attempt to gain access to a system.
Recent research within the cybersecurity analytics group has been studying the relationship between these kinds of attack surfaces and the kinds of harm that an organisation might be exposed to. An interesting question would be whether an orientation around intent, or harm, might result in a different test strategy; would a different focus be given to the kinds of attack vectors explored in a test if a particular harm is aimed at.
This mini-project would aim to explore this question by designing penetration test strategies based on a set of particular harms, and then seek to consider potential differences with current penetration practices by consultation with the professional community.
Students will need to have a working understanding of penetration testing techniques. The purpose of this project is to implement one or more photogrammetry techniques from a series of 2D photographs.
The student is free to approach the challenge as they see fit, but would be expected to design, implement and assess the tool they develop. Combined, these RTI images form a single photograph in which users can relight these objects by moving the light sources around the hemisphere in front of the object, but also specify user-defined parameters, including removing colour, making the objects more specular or diffuse in order to investigate the surface details in depth.
It can be used for forensic investigation of crime scenes as well as cultural heritage documentation and investigation. The purpose of this project is to implement RTI methods of their preference. This project will seek to develop a library of such trip-wires based on a survey of openly documented and commented upon attacks, using the Oxford framework. There will be an opportunity to deploy the library into a live-trial context which should afford an opportunity to study the relative utility of the trip-wires within large commercial enterprises.
For instance, many model checking problems can be reduced to the solution of parity games. The exact complexity of these games is unknown: However, in practice, most parity games are solved efficiently. This project aims at defining new algorithms for parity games and evaluating them in practice, in particular, over random graphs.
Computing Hilbert bases is a fundamental problem encountered in various areas in computer science and mathematics, for instance in decision procedures for arithmetic theories, the verification of infinite-state systems and pure combinatorics.
With the ubiquity of multi-core architectures, it seems conceivable that, with proper engineering, this approach will outperform existing approaches. The goal of this project is to deliver an implementation of the algorithm of [1] and benchmark it against competing tools. If successful, an integration into the SageMath platform could be envisioned. This theory is a strict fragment of Presburger arithmetic, an arithmetic theory that finds numerous applications, for instance, in the verification of infinite-state systems.
If time permits, an implementation of the decision procedure developed in this project could be envisioned. The key insight is that this noise injection prevents the learnt weights from being too delicately balanced to fit the data; some kind of robustness is necessary to fit noisy data.
Another interesting consequence of noisy data is that recent work shows that learning algorithms using noisy data may be better at protecting privacy of the data. Thus, there may be twin advantages to this approach.
This project will involve understanding the backgroud in this topic, performing simulations to understand the behaviour and hopefully developing new theories. This project may involve collaboration with Mr. Alexis Poncet and Dr.
There has been much recent work in understanding evolution through a computational lens. One of the fundamental building blocks of life is circuits where the production of protein is controlled by other ones transcription factors ; these circuits are known as transcription networks. Mathematical models of transcription networks have been proposed using continuous-time Markov processes.
The focus of the project is to use these models to understand the expressive power of these networks and whether simple evolutionary algorithms, through suitably guided selection, can result in complex expressive patterns. The work will involve both simulations and theory. The project will involve a survey of past methods used, as well as data collection and model fitting.
This project is open-ended and hence potentially risky ; the main aim is to get good results using historical and current data. It would be helpful if the student has good programming experience as wel as knowledge of various different machine learning techniques.
The project will involve collabortion with Dr. This work will build upon the methods presented within the first year Linear Algebra and Continuous Maths courses. Design and computational implementation of fast and reliable numerical models arising from biological phenomena. The Square Root Law of Steganography: Steganalysis is the art of detecting that hiding took place.
A key question is how the amount of information that can be securely hidden i. In I co-authored a paper showing that my theoretical "square root law" was observed experimentally, using state-of-the-art for hiding and detection methods.
This project is to run similar experiments using methods 10 years more modern. It would involve combining off-the-shelf code some in MATLAB, some in Python from various researchers and running fairly large scale experiments to measure detection accuracy versus cover size and payload size in tens of thousands of images, then graphing the results suitably. Ability to piece together others' code and draw graphs nicely.
It is necessary to have some experience working with video codecs, for example experience contributing to the H. Most pure research focuses on bitmap or JPEG images, or simple video codes.
In practice, there are often further constraints: This project is primiarly programming. First we must characterize the properties of image transmission through Twitter, and then provide an implementation of image steganography through it. This might be a complete piece of software with a nice interface, if the channel is straightforward, or a proof-of-concept if the channel causes difficult noise in which case we will need suitable error-correcting codes. It would be useful to know a little about the JPEG image format before starting the project.
The most effective ways to detect steganography are machine learning algorithms applied to "features" extracted from the images, trained on massive sets of known cover and stego objects. The images are thus turned into points in high-dimensional space. We have little intuition as to the geometrical structure of the features do images form a homogeneous cluster? This is a programming project that creates a visualization tool for features, extracting them from images and then projecting them onto 2-dimensional space in interesting ways, while illustrating the effects of embedding.
Linear Algebra, Computer Graphics. Machine Learning an advantage. The long-term objective of this project is to provide a reusable interface where future students can pit their game-playing engines against each other. The focus here is on good object-oriented design, reusability, easy-to-use interfaces for game engines, and an appealing graphical user interface.
The student should also create two possibly simple engines to test the software. Time permitting, those engines might involve advanced techniques in order to play better. In a previous project, an interface was created to allow future students to pit their game-playing engines against each other.
In this project the goal is to program a strong Hex engine. The student may choose the algorithms that underly the engine, such as alpha-beta search, Monte-Carlo tree search, or neural networks.
The available interface allows a comparison between different engines. It is hoped that these comparisons will show that the students' engines become stronger over the years. The project involves reading about game-playing algorithms, selecting promising algorithms and datastructures, and design and development of software in Java or Scala. The goal of this project is to explore if the number of columns of M can be chosen below 11, perhaps by dropping some columns of M.
This will require some geometric reasoning. The project will likely involve the use of tools such as SMT solvers that check systems of nonlinear inequalities for solvability. The project is mathematical in nature, and a good recollection of linear algebra is needed.
Openness towards programming and the use of tools is helpful. An algorithm is incentive compatible when the participants have no reason to lie or deviate from the protocol. In general, we have a good understanding when the problem involves only buyers or only sellers, but poor understanding when the market has both buyers and sellers. This project will investigate and implement optimal algorithms for this problem.
It will also consider algorithms that are natural and simple but may be suboptimal. Mathematical and algorithmic maturity. The main difficulty that the bitcoin protocol tries to address is to achieve agreement in a distributed system run by selfish participants. To address this difficulty, the bitcoin protocol relies on a proof-of-work scheme. Since the participants in this proof-of-work scheme want to maximize their own utility, they may have reasons to diverge from the prescribed protocol.
The project will address some of the game-theoretic issues that arise from the bitcoin protocol Prerequisites: There are efficient algorithms for LP, such as the Simplex Algorithm, that do not perform well in the worst case, and there are inefficient algorithms, such as the Ellipsoid Algorithm, that are good in theory but not in practice. Learning algorithms, such as the Multiplicative Weight Update Algorithm, have the potential to be good both in theory and in practice.
The project will address this question with theoretical analysis and implementation. Below are some concrete project proposals: Modelling trust in human-robot interaction. When human users interact with autonomous robots, appropriate notions of computational trust are needed to ensure that their interactions are safe and effective: Trust management systems have been introduced for autonomous agents on the Internet, but need to be adapted to the setting of mobile robots, taking into account intermittent connectivity and uncertainty on sensor readings.
Recently, a logic and model checking algorithm were formulated for reasoning about trust http: This project aims to develop an implementation of a simplified model checking algorithm. This project is concerned with synthesising strategies for autonomous driving directly from requirements expressed in temporal logic, so that they are correct by construction. Probability is used to quantify information about hazards, such as accidents hotspots.
The idea is to develop the techniques further, to allow high-level navigation based on waypoints, and to develop strategies for avoiding threats, such as road blockage, at runtime. In the longer term, the goal is to validate the methods on realistic scenarios in collaboration with the Mobile Robotics Group.
The project will suit a student interested in theory or software implementation. For more information about the project see http: Autonomous robots have numerous applications in scenarios such as warehouse management, planetary exploration, or search and rescue. In view of environmental uncertainty, such scenarios are modelled using Markov decision processes.
This project aims to develop a PRISM model of a system of robots for a particular scenario so that safety and effectiveness of their cooperation is guaranteed. This project will suit a student interested in machine learning and software implementation.
Modelling and verification of DNA programs. DNA molecules can be used to perform complex logical computation. For more information about the DNA computing project see http: Safety Testing of Deep Neural Networks. Despite the improved accuracy of deep neural networks, the discovery of adversarial examples has raised serious safety concerns.
In a recent paper https: The method is based on a two-player turn-based stochastic game, where the first player's objective is to find an adversarial example by manipulating the features, and proceeds through Monte Carlo tree search. It was evaluated on various networks, including YOLO object recognition from camera images. This project aims to adapt the techniques to object detection in lidar images such as Vote3D http: It inputs camera images and produces a steering angle. The network is trained on data from cars being driven by real drivers, but it is also possible to use the Udacity simulator to train it.
Safety concerns have been raised for neural network controllers because of their vulnerability to adversarial examples — an incorrect steering angle may force the car off the road. This project aims to use these techniques to evaluate the robustness of PilotNet to adversarial examples. Since deep neural networks DNNs are deployed in autonomous driving systems, ensuring their safety, security and robustness is essential.
Unfortunately, DNNs are vulnerable to adversarial examples - slightly perturbing an image may cause a misclassification, see CleverHans https: Such perturbations can be generated by adversarial attackers. Most current adversarial perturbations are designed based on the L1, L2 or Linf norms. Recent work demonstrated the advantages of perturbations based on the L0-norm http: This project, aims to, given a well-trained deep neural network, demonstrate the existence of a universal image-agnostic L0-norm perturbation that causes most of images to be misclassified.
The work will involve designing a systematic algorithm for computing universal perturbations and empirically analysing these to show whether they generalize well across neural networks. This project is suited to students familiar with neural networks and Python programming.
Security analysts often find it necessary to perform n Network scanning, probing and vulnerability testing aids the process of discovering and correcting network vulnerabilities. There exist a number of challenges in terms of gathering network data in this manner: A number of approaches — such as that by Roschke et al.
However, the databases return the results in different formats - some in textual format, others in XML format. Scanning and then combining the results from multiple scanners takes a lot of time Cheng et al. Quite often the network configuration changes during the scan — which means that the results are quite often inaccurate. This thesis addresses one or more of the challenges presented above. This project involves practical work which may involve setting up a virtual network and applying a range of scanning, probing and vulnerability testing mechanisms on the network.
This research analyses the problem of calculating network security metrics and proposes a framework which can calculate security metrics for a typical small network comprising of numerous devices, operating systems and hosts. The project will involve the configuration of virtual networks for testing the framework. CVSS or other metric scores may be used to aid the calculation of the metric.
This dissertation involves some practical work which may involve setting up a virtual network which has a number of machines on the network and applying a range of scanning, probing and vulnerability testing mechanisms on the network. This is in contrast to typical programming paradigms, wherein programmers explicitly spell out procedures for carrying out the computation. Constraint programming has a wide range of applications e.
SAT-solvers claimed by some researchers to be among the greatest achievements in the past decade are one of the most powerful solvers in the constraint programming toolbox.
In a nutshell, SAT-solvers are algorithms for solving satisfiability of boolean formulas, which is an NP-complete problem that can be used as a universal language for encoding many practical combinatorial problems.
The aim of the project is to investigate ways in which to improve the performance of SAT-solvers by embedding symmetry information e. Boolean formulas that arise in practice e. Specific questions to explore include: The concurrent datatypes could be some of those studied in the Concurrent Algorithms and Datatypes course. Typical properties to be proved would be linearizability against a suitable sequential specification and lock freedom.
Concurrency and Concurrent Algorithms and Datatypes. I recently proposed an approach to this problem [1]. The idea is to build a process that: The Concurrency course would be a prerequisite. Unfortunately, the definition is convoluted and hard to understand. The aim of this project is to aid better understanding of the JMM, by producing a tool that, given a small program like the one above or the ones in [4] , returns all its valid executions.
Understand the requirements of e-commerce protocols; Specify an e-commerce protocol, both in terms of its functional and security requirements; Understand cryptographic techniques; Understand how these cryptographic techniques can be combined together to create a secure protocol - and understand the weaknesses that allow some protocols to be attacked; Design a protocol to meet the requirements identified; Implement the protocol.
This evaluation aims to provide deeper insights into the behavior of these hashmaps under different types of workloads using performance profiling tools e. Armed with this knowledge, the student will implement a hashmap with multi-index access optimised for using in real-time streaming environments. In this project, we consider one such state-of-the-art engine, called DBToaster, which continuously evaluates static queries over changing dynamic data.
The goal of this project is to enable multi-core query processing over streaming data in DBToaster, which requires implementing necessary primitives for parallelization of the existing single-threaded query evaluation procedures. The project will contribute to such implementations. Prerequisites Interest in open source software, some knowledge of Python, appropriate maths background.
Babai and enjoyed attention in machine learning community recently cf. While it is relatively easy to implement, efficient portable implementations seem to be hard to find. In this project we will work on producing such an open-source implementation, either by reworking our old code from http: It works by delivering small doses of tracer gases into the patient breaths and measuring the responses in the expired breaths.
The device is being developed towards commercialisation. A lung simulation that could be used by non-expert computer scientists such as nurses and medical doctors would be a useful addition to the technology.
To develop a user-friendly lung simulation to help predicting the responses of the inspired sinewave tests in various lung conditions, from healthy to diseased.
The simulation consists of 2 key parts: Details of the mathematical lung model are available up on request. The model to be fitted might be but is not limited to a system of Ordinary Differential Equations and the Bayesian estimation tools might be build around an existing system such as Stan, PyML or Edward.
A good tutorial system should be able to let the user change the underlying model system, introduce noise to a system, visualise interactive updates to probability distributions, explore the progress of a chosen sampling method such as Metropolis-Hastings and provide enough information that a novice student can get an intuition into all aspects of the process.
Computer graphics, Object-oriented programming The idea behind this project is to build an educational tool which enables the stages of the graphics pipeline to be visualised. For example the Gnome disk usage analyzer Baobab uses either a "ring chart" or "treemap chart" Representation to show us which sub-folders are using the most disk. In the early s the IRIX file system navigator used a 3D skyscraper representation to show us similar information.
There are plenty more ways of representing disk usage: What kind of representation is most intuitive for finding a file which hogging disk-space and which is most intuitive for helping us to remember where something is located in the file-system tree?
The aim is to explore other places where visualisation gain intuition: In many current non-linguistic applications, however, this kind of "what action do I perform next" question is answered by using reinforcement learning, where the system learns a reward function from training data that should bias towards that action most likely to lead to a successful conclusion. This project aims to experiment to see whether a reinforcement learning decision component could lead to better parsing performance than the more usual classifier-based decisions.
They can be used to explore real hardware implementations of processor designs from simple accumulator machines, through to register machines, and more unusual stack machines. Their use outside industry has previously been limited due to the high cost and complexity of their associated development software. However, this is changing, with an open-source toolchain for popular FPGA devises becoming available in the last year the equivalent of Linux in the OS world.
This project will build on this toolchain to develop the additional software necessary to allow students to design and explore simple processor designs on a custom FPGA development board. The current toolchain requires the use of the Verilog hardware description language. Verilog is very powerful but also very general. This project will develop a high-level language possibly a graphical language focused on the development of simple processor designs from a small number of standard components such as RAM, multiplexers and registers.
While training such networks is computationally expensive typically requiring very large image datasets and exploiting GPU acceleration , they can often be deployed on much simpler hardware if simplifications such as integer or even binary weights are imposed on the network.
This project will explore the deployment of trained convolution networks on microcontrollers and possibly also FPGA-based hardware with the intention of demonstrating useful image processing perhaps recognising the presence of a face in the field of view of a low pixel camera on low-power devices.
NASA's Mars Climate Orbiter was lost in due to software that calculated trajectory thruster firings in pounds seconds, rather than newton-seconds. In other cases, dimensional analysis statically checking that the computed units match those that are expected is sufficient to catch many errors in calculations.
While F supports units natively, and libraries exist in many others languages e. Java, Haskell, Python , none are particularly easy to use, and often introduce clumsy syntax. This project will build improve on these approaches. You will be required to develop a unit-aware interactive programming environment enabling unit-safe physics based calculations to be performed.
This might be a stand-alone solution, or a kernel for an existing interactive computing environment such as Project Jupyter jupyter. This project is not available to MSc students in B This project is not available to MSc students in The combination leads to poor learning outcomes, low engagement, dissatisfaction and high dropout rates. To build this system, the following modules will need to be developed: To read more about the role of machine learning in education — see medianetlab.
This is a hard problem, since the preferences that agents report might contradict each other, and this leads to so-called voting paradoxes. Also, it can be computationally hard to calculate what decisions to make. A promising way to tackle this problem is by exploiting structure in the reported preferences. For this purpose, Australian researchers have collected an impressive amount of real-world preference data PrefLib http: This project is about analysing this data to reveal how much structure is contained in these preferences.
There are different measures of closeness, and for many of them the associated decision problem is NP-hard; for others, the computational complexity is not known. There are several ways in which this project can be pursued. On the one hand, one can consider the notions of closeness for which the complexity of the associated problem is not known, and try to develop an efficient algorithm or prove NP-hardness.
Machine learning is increasingly used in finance to make predictions as well as to aggregate among existing strategies for making investments over time. We will use various free as well as proprietary data sets to assess the value of our newly developed methods in terms of both profit and risk, and compare them with state of the art techniques.
The expectation is that the work will lead to a conference publication. Federico Leardini , il toccante ricordo dei colleghi nel servizio di Le pagelle di Sampdoria - Torino: Come vedere il match in tv e in Super Bowl , come vederlo in tv o in streaming.
Come vedere Udinese - Milan , in tv o in streaming Udinese Milan in diretta: Juventus Sassuolo in diretta: Juventus - Sassuolo , formazioni ufficiali e diretta. J - Ax e Fedez in lacrime per la fan Denise.
La rivelazione di Filippa Lagerback: Samantha De Grenet a Verissimo in lacrime: Verissmo, Samantha De Grenet non riesce a trattenere le lacrime in Usa, padre di tre vittime aggredisce Larry Nassar. Rilasciato, diventa un eroe del web Larry Nassar aggredito in tribunale.