. "Computer Science"@en . . "English"@en . . "Mathematics"@en . . "Mechanical engineering"@en . . "Introduction to machine learning"@en . . "6.7" . "The goal of this course is to provide an accelerated preparation for the more advanced courses in the Michaelmas term. This preparation includes several foundational topics in machine learning and through these we cover essential mathematical background and optimisation tools.\r\n\r\nAims:\r\n\r\nProvide a thorough introduction into the topic of statistical inference including maximum-likelihood and Bayesian approaches. Introduce important tools from probability and statistics.\r\nIntroduce algorithms for regression, classification, clustering and sequence modelling. Through these use mathematical tools including linear algebra, eigenvectors and eigenvalues, multidimensional calculus, and calculus of variations. \r\nIntroduce basic concepts in optimisation and dynamic programming, including gradient ascent and belief propagation. \n\r\nObjectives:\r\n\r\nUnderstand the use of maximum-likelihood and Bayesian inference and the strengths and weaknesses of both approaches. \r\nImplement methods to solve simple regression, classification, clustering and sequence modelling problems. \r\nImplement simple optimisation methods (gradient and coordinate descent, EM) and dynamic programming (Kalman filter or Viterbi decoding). \n\nOutcome: Not Provided" . . "Presential"@en . "TRUE" . . "Speech recognition"@en . . "3.4" . "Aims\r\n\r\nThe aim of this module is to introduce the issues in speech recognition and discuss the statistical and deep learning approaches used to build automatic speech recognition (ASR) systems.\n\nOutcome:\nOn completion of this module, students should understand:\r\n\r\nhidden Markov acoustic models, N-gram language models, and their use in speech recognition\r\nthe use of various neural network acoustic models\r\nhow large vocabulary speech recognition operates\r\nfeature extraction and processing \r\ntechniques for adaptation\r\ndiscriminative sequence training procedures\r\nend-to-end trainable speech recognition approaches." . . "Presential"@en . "TRUE" . . "Computer vision"@en . . "6.7" . "Aims\r\n\r\nThe aims of the course are to:\r\n\r\nintroduce the principles, models and applications of computer vision;\r\ncover image structure, projection, stereo vision, structure from motion and object detection and recognition;\r\ngive case studies of industrial (robotic) applications of computer vision, including visual navigation for autonomous robots, robot hand-eye coordination and novel man-machine interfaces.\n\nOutcome:\nAs specific objectives, by the end of the course students should be able to:\r\n\r\ndesign feature detectors to detect, localise and track image features;\r\nmodel perspective image formation and calibrate single and multiple camera systems;\r\nrecover 3D position and shape information from arbitrary viewpoints;\r\nappreciate the problems in finding corresponding features in different viewpoints;\r\nanalyse visual motion to recover scene structure and viewer motion, and understand how this information can be used in navigation;\r\nunderstand how simple object recognition systems can be designed so that they are independent of lighting and camera viewpoint;\r\nappreciate the industrial potential of computer vision but understand the limitations of current methods.\r\nLectures for this course are shared with 4th year Engineering undergraduates." . . "Presential"@en . "TRUE" . . "Deep learning & structured data"@en . . "6.7" . "Aims\r\nThe aims of the course are to:\r\n\r\nteach the basic concepts of deep learning and forms of structure that can be used for generative and discriminative models. In addition, the use of models for classifying structured data, such as speech and language, will be discussed\n\nOutcome:\nObjectives\r\nAs specific objectives, by the end of the course students should be able to:\r\n\r\nunderstand the basic principles of pattern classification and deep learning;\r\nunderstand generative and discriminative models for structured data;\r\nunderstand the application of deep-learning to structured data;\r\napply pattern processing techniques to practical applications.\r\n* This module is shared with 4th Year undergraduates from the Department of Engineering." . . "Presential"@en . "TRUE" . . "Probabilistic machine learning"@en . . "6.7" . "Aims\r\n \r\nThe aim of the course is to:\r\nintroduce students to basic concepts in machine learning, focusing on statistical methods for supervised and unsupervised learning.\n\nOutcome:\nObjectives\r\n \r\nBy the end of the course students should be able to:\r\ndemonstrate a good understanding of basic concepts in statistical machine learning;\r\napply basic machine learning methods to practical problems.\r\n \r\nThis module is shared with 4th year undergraduate students from the Department of Engineering." . . "Presential"@en . "TRUE" . . "Advanced speech recognition"@en . . "2" . "No Description, No Learning Outcome" . . "Presential"@en . "TRUE" . . "Neural machine translation and dialogue systems"@en . . "2" . "Aims\n\nThis half-module provides an introduction to machine translation and task-oriented dialogue systems as problems that can be addressed by machine learning. The presentation will employ sequence-to-sequence models to develop a uniform approach to these problems.\n\nOutcome:\nOn completion of this model, students should have a working familiarity with:\r\n\r\ntranslation and dialogue as problems in natural language processing;\r\ndata sets used in creating dialogue systems and machine translation systems;\r\nautomatic and manual assessment of dialogue and translation quality;\r\nthe statistical approach to task oriented dialogue systems and its component tasks;\r\nmodelling approaches for neural machine translation;\r\nsequence-to-sequence models, such as the Transformer architecture and instances such as GPT2\r\nfine tuning and domain adaptation procedures;\r\ncurrent research problems, including search and model correctness\r\ndata biases and ethical concerns in translation and dialogue." . . "Presential"@en . "TRUE" . . "Spoken language generation and processing"@en . . "2" . "Aims\n\nThe aim of this module is to teach students the underlying concepts and examples of spoken language processing. This module extends the speech recognition module to examine state-of-the-art systems, and how these systems are connected to downstream applications. Finally the module discusses speech synthesis systems to enable system feedback to users.\n\nOutcome:\nOn completion of this module, students should:\r\n\r\nUnderstand state-of-the-art speech recognition systems\r\nUnderstand the challenges of integrating speech recognition and language processing\r\nBe able to develop (from examples) spoken language processing systems \r\nUnderstand speech synthesis systems" . . "Presential"@en . "TRUE" . . "Advanced machine learning"@en . . "5" . "Aims\r\n\r\nThe aim of this module is to teach advanced topics that will enable students to follow state-of-the-art research in machine learning.\n\nOutcome:\nOn completion of this module, students should:\r\n\r\nunderstand advanced topics in machine learning;\r\nbe able to read current research papers in the field\r\nbe able to implement state of the art learning algorithms\r\nbe ready to conduct research in the field." . . "Presential"@en . "TRUE" . . "Reinforcement learning and decision making"@en . . "2" . "Aims\r\n\r\nThis module introduces basic principles of sequential decision making under uncertainty and the application in Reinforcement Learning and Control. Foundations and recent algorithms are covered.\n\nOutcome:\r\nOn completion of this module, students should understand:\r\n\r\nThe foundations of sequential decision making and reinforcement learning\r\nThe connections between control and reinforcement learning\r\nThe exploration vs exploitation trade-off." . . "Presential"@en . "TRUE" . . "Designing intelligent interactive systems"@en . . "2" . "Aims:\r\n\r\nThis half-module aims to: \r\n\r\nProvide a basic understanding of design and human-computer interaction theories and methods. \r\nIntroduce a systematic process for designing intelligent interactive systems. \r\nIntroduce design tactics for realising effective intelligent interactive systems.\n\nOutcome:\nBy the end of the module, students should be able to:\r\n\r\nApply a systematic design engineering process to design an interactive system.\r\nApply qualitative and quantitative methods to gain an understanding of users’ needs and wants.\r\nUnderstand elementary human behavioural theory, as it applies to user interface design.\r\nModel user behaviour and understand the limitations and implications of such modelling.\r\nUnderstand design strategies for interactive systems that infer or predict user behaviour.\r\nUnderstand the role of verification and validation and have basic knowledge of common verification and validation strategies for interactive systems." . . "Presential"@en . "TRUE" . . "Natural language processing"@en . . "5" . "Aims:\r\n\r\nThis course introduces the fundamental techniques of natural language processing. It aims to explain the potential and the main limitations of these techniques. Some current research issues are introduced and some current and potential applications discussed and evaluated. Students will also be introduced to practical experimentation in natural language processing.\n\nOutcome:\nOn completion of this module, students should:\r\n\r\nbe able to discuss the current and likely future performance of several NLP applications;\r\nbe able to describe briefly a fundamental technique for processing language for several subtasks, such as morphological processing, parsing, word sense disambiguation etc.;\r\nunderstand how these techniques draw on and relate to other areas of computer science;\r\nunderstand the basic principles of designing and running an NLP experiment.\r\nLectures for this course take place in the Department of Computer Science and Technology. Other elements of the course are taken in the Department of Engineering." . . "Presential"@en . "FALSE" . . "Practical optimisation"@en . . "5" . "Aims\r\n\r\nThe aims of the course are to:\r\n\r\nTeach some of the basic optimisation methods used to tackle difficult, real-world optimisation problems.\r\nTeach means of assessing the tractability of nonlinear optimisation problems.\r\nDevelop an appreciation of practical issues associated with the implementation of optimisation methods.\r\nProvide experience in applying such methods on challenging problems and in assessing and comparing the performance of different algorithms.\n\nOutcome:\nAs specific objectives, by the end of the course students should be able to:\r\n\r\nUnderstand the basic mathematics underlying linear and convex optimisation.\r\nBe able to write and benchmark simple algorithms to solve a convex optimisation problem.\r\nUnderstand the technique of Markov-Chain Monte Carlo simulation, and apply it to solve a Travelling Salesman Problem.\r\nUnderstand the ways in which different heuristic and stochastic optimization methods work and the circumstances in which they are likely to perform well or badly.\r\nUnderstand the principles of multi-objective optimization and the benefits of such of approaching real-world optimization problems from a multi-objective perspective.\r\n* This module is shared with 4th Year undergraduates from the Department of Engineering." . . "Presential"@en . "FALSE" . . "Advanced robotics"@en . . "5" . "Aims:\r\n\r\nThe aims of the course are to:\r\n\r\nLearn advanced topics of robotics (underactuated robotics, robot learning, soft robotics, human robot interactions, and distributed robotics)\r\nFundamentals (theories and methodologies) of advanced robotics researches\r\nPractical implementation of advanced robotics technologies\n\nOutcome:\nAs specific objectives, by the end of the course students should be able to:\r\n\r\nExtend the knowledge of introductory robotics to more advanced ones to carry out research\r\nLearn research techniques and skills for robotics projects\r\nWork effectively with collaborators in robotics projects\r\nDeliver professional presentations and communication of robotics projects\r\n* This module is shared with 4th Year undergraduates from the Department of Engineering." . . "Presential"@en . "TRUE" . . "Introduction to robotics"@en . . "5" . "Aims\r\n\r\nThe aims of the course are to:\r\n\r\nIntroduce fundamentals of robotics\r\nLearning technologies and techniques to design, assemble, and control robots\r\nHands-on exercises on robot development through projects\r\nPresentation of research and development\n\nOutcome:\nAs specific objectives, by the end of the course students should be able to:\r\n\r\nLearning different design strategies and architectures of robots\r\nDesign methods of automated complex systems\r\nDevelopment of simulated complex robots\r\nModel-based analysis robot performance\r\n* This module is shared with 4th Year undergraduates from the Department of Engineering." . . "Presential"@en . "FALSE" . . "Machine learning and the physical world"@en . . "5" . "Aims\r\n\r\nThe module “Machine Learning and the Physical World” is focused on machine learning systems that interact directly with the real world. Building artificial systems that interact with the physical world have significantly different challenges compared to the purely digital domain. In the real world data is scares, often uncertain and decisions can have costly and irreversible consequences. However, we also have the benefit of centuries of scientific knowledge that we can draw from. This module will provide the methodological background to machine learning applied in this scenario. We will study how we can build models with a principled treatment of uncertainty, allowing us to leverage prior knowledge and provide decisions that can be interrogated.\r\n\r\nThere are three principle points about machine learning in the real world that will concern us.\r\n\r\nWe often have a mechanistic understanding of the real world which we should be able to bootstrap to make decisions. For example, equations from physics or an understanding of economics.\r\nReal world decisions have consequences which may have costs, and often these cost functions need to be assimilated into our machine learning system.\r\nThe real world is surprising, it does things that you do not expect and accounting for these challenges requires us to build more robust and or interpretable systems.\r\nDecision making in the real world hasn’t begun only with the advent of machine learning technologies. There are other domains which take these areas seriously, physics, environmental scientists, econometricians, statisticians, operational researchers. This course identifies how machine learning can contribute and become a tool within these fields. It will equip you with an understanding of methodologies based on uncertainty and decision making functions for delivering on these challenges.\n\nOutcome:\nYou will gain detailed knowledge of\r\n\r\nsurrogate models and uncertainty\r\nsurrogate-based optimization\r\nsensitivity analysis\r\nexperimental design\r\nYou will gain knowledge of\r\n\r\ncounterfactual analysis\r\nsurrogate-based quadrature\r\n* Lectures take place in the Department of Computer Science and Technology and are part of the MPhil in Advanced Computer Science." . . "Presential"@en . "FALSE" . . "Software engineering and design"@en . . "5" . "No Description\n\nOutcome:\nAs specific objectives, by the end of the course students should be able to:\r\n\r\nUnderstand the benefits of object-oriented analysis and design, its concepts and processes.\r\nBe familiar with formal design tools for object orientated design and analysis.\r\nRecognise and understand some frequently used design patterns.\r\nBe aware of the process involved in user interface design.\r\nUnderstand software development methodologies.\r\nUnderstand the main issues and processes necessary to achieve effective software product development.\r\nBe familiar with main challenges of software innovation and the strategies and opportunities to address them.\r\n* This module is shared with 4th Year undergraduates from the Department of Engineering." . . "Presential"@en . "TRUE" . . "Research project"@en . . "5" . "In addition to the taught course modules, students will undertake a research project leading to a dissertation and a poster presentation.\r\n\r\nProjects are formulated and planned during Lent Term and there is no need to contact supervisors ahead of time. A list of project options will be provided to students. These will be designed to ensure that the project is scoped correctly so that students can complete it with the time, data, and computing resources available. The assumption is that students will choose from project topics proposed by the Faculty, however students with particular research interests will have the option of working with a member of staff to design and propose their own topic (with the approval of the Course Director).\r\n\r\nEach student will have a project supervisor, and the project topics will be approved by the course management committee. The project is then carried out over the Easter and Research terms. Students will be expected to attend the Computational and Biological Learning and Speech Group programmes of research seminars. The MPhil students will be integrated into the research groups in the Department and will work closely with PhD students and postdocs under the direction of the project supervisor. Projects will be evaluated on the basis of a dissertation of up to 15,000 words and a poster presentation.\n\nOutcome: Not Provided" . . "Presential"@en . "TRUE" . . "Master in Machine Learning and Machine Intelligence"@en . . "https://www.mlmi.eng.cam.ac.uk/" . "60"^^ . "Presential"@en . "The goal of the field of Machine Intelligence is to develop systems that can perceive the world, plan and make decisions, interact with humans and other intelligent agents, and provide explanations for their actions. Machine Learning provides many of the technical tools used to develop intelligent systems. This field overlaps with statistics and computer science.\r\n\r\nThe MPhil in Machine Learning and Machine Intelligence is an eleven month full-time programme offered by the Machine Learning Group, the Speech Group, and the Computer Vision and Robotics Group in the Cambridge University Department of Engineering. The course aims to teach the state-of-the-art in machine learning, speech and language processing, and computer vision; to give students the skills and expertise necessary to take leading roles in industry and to equip them with the research skills necessary for doctoral study at Cambridge and other universities.\n\nOutcome:\nThe MPhil in Machine Learning and Machine Intelligence offers many opportunities for the development of professional skills, giving students experience in preparing and giving presentations, report writing, collaborating in research teams, and carrying out literature searches. Students will also be expected to attend the invited seminar series run by the Machine Learning and the Speech Group."@en . . . . "1"@en . "FALSE" . . "Master"@en . "Thesis" . "17022.00" . "British Pound"@en . "41694.00" . "None" . "Employment prospects are also extremely good for students who plan to go directly into industry. The course will impart directly employable skills and expertise which are in great demand in the IT, financial, and manufacturing sectors."@en . "4"^^ . "TRUE" . "Midstream"@en . . . . . . . . . . . . . . . . . . . . "School of Technology - Department of Engineering"@en . .