Merton Presentations 2018


Week 3, Saturday: Deadline for titles and abstracts submissions. 

Week 6, Monday (Breakfast Room) 09:30-12:00 and 14:30-17:00 : Rehearsals. 

Monday Week 6
09:30 - 10:00 Leonie Woodland
10:00 - 10:30Georgia Acton
10:30 - 11:00Jack McIntyre
11:00 - 11:30Catherine Felce
11:30 - 12:00Roshan Dodhia
14:30 - 15:00 Jacob Robertson
15:00 - 15:30 Andrew Doyle
15:30 - 16:00Lucy Biddle
16:00 - 16:30Jules Desai

Week 8, Monday 16:00-20:00 (Ian Taylor Room): Final presentations.

Titles and Abstracts

Lucy Biddle

Title: The Science of Storage

Abstract: Computers today bear very little resemblance to those from 50 years ago, and much of this change is due to the development of increasingly space-efficient forms of data storage and processing. In this talk I will present the evolution of computer hardware, explain the limitations faced by current technology and outline what the next generation of storage could look like.

Jules Desai

Title: Modern Art and Modern Physics: Painting the Paradigm Shift

Abstract: The early twentieth century saw physics turned on its head by the likes of Einstein, Planck and Schrödinger. Along with this came a philosophical crisis as the world attempted to understand the epistemic and metaphysical implications of this “Modern Physics”.  At the same time, artists began to change their perspective on the function and purpose of their art. Art was no longer about the about the narrative but instead moved towards the abstract and the unconscious reaction. This talk will investigate how the paradigm shifts in modern physics influenced and shaped the art of its time. I will consider how, in their work, artists attempted to embody concepts such as non-Euclidean geometry, higher dimensionality and relative simultaneity. I hope to convince you that art and physics are more tightly intertwined than you might think (but, after all, the more one thinks about this the more evident it becomes!).

Jacob Robertson

Title: Random Processes in Genetics

Abstract: In this talk I will discuss a stochastic model of evolution with the aim of calculating certain interesting quantities associated with genetic drift. The talk will begin with a brief discussion of natural selection and the theoretical need for discrete units of heredity, of exactly the kind Gregor Mendel had already experimentally found. The rest of the talk focusses on the Moran model, a stochastic model of evolution, in the absence of mutation or selective effects. This will enable a calculation of the probability that a certain allele ‘reaches fixation’ i.e. obtains a frequency of 100% in the population as well as an estimate for the time it will take to do so. The talk concludes by showing how in the limit of large populations the Moran model becomes a diffusive process. 

Andrew Doyle

Title: Neural Networks and Machine learning

Artificial intelligence is evolving rapidly. Whether it will be self-driving cars or almost human robots, nobody knows exactly how it will impact our lives, but the immense power and potential of it are quickly becoming apparent. In this talk I plan to outline the basics of how Neural Networks are structured and go on to explain how machines “learn” by showing a common example of hand written digit recognition. Finally I will look at current projects that may shape our future.

Roshan Dodhia

Title: What does it mean to own a bitcoin?

Abstract: Bitcoin is a cryptocurrency and worldwide payment system. Since it’s creation in 2009, there has been a growing interest in this mysterious idea of ‘mining’ for bitcoins. We are all very familiar with a concept of digital currency in the form of credit and debit cards, but few people really understand what it means to own a bitcoin. In my talk, I will explain how cryptocurrencies like Bitcoin work, why they are different from normal currencies we are so comfortable with and finally I’d like to tackle the million dollar (or 114.17 bitcoin) question: should you be investing in cryptocurrencies?

Georgia Acton

Title: The Science Behind Hubble

Abstract: Since its launch in 1990 The Hubble Space Telescope has contributed to many ground-breaking discoveries; from observations that helped to calculate the rate of expansion of the universe, to mapping the 3-D distribution of dark matter, Hubble has aided the world of science in many ways. In this presentation I want to explore the science that made Hubble possible and the technology that has led to its extraordinary observations. I will discuss the camera technology on-board, the science and need for a thermoelectric cooler, the presence of a spectrograph, and the use of gyroscopes and flywheels to orientate and focus itself. Hubble has not only help uncover amazing science, it is itself a beautiful piece of technology and I hope to convey this throughout.

Jack McIntyre

Title: Arrow’s Impossibility Theorem: The trouble with ranked voting

Abstract: Imagine that you are tasked with designing a voting system. Each voter gets to rank a group of candidates from best to worst and you must somehow turn these individual rankings into a single ranking for the whole electorate. This should be done in such a way that gives the same output ranking each time that your system is presented with the same set of voter preferences. A potential first step would be to decide on some criteria that your system should fulfil to be considered fair. One thing that you may suggest is that if all the voters prefer candidate A to candidate B then the ranking for the group should also prefer A to B. It would also seem sensible to require that whether candidate A beats candidate B or candidate B beats candidate A should be independent of how the voters rank candidate C. It turns out that the only way to satisfy both of these, seemingly reasonable, criteria is to have what is known as a dictatorship, where the result of the election is completely determined by how one particular voter fills in their ballot. This rather surprising result is know as Arrow’s impossibility theorem. In this presentation I will prove the theorem and discuss where it leaves ranked voting.

Leonie Woodland

Title: The Physics and Physiology of Vowel Production and Perception

Abstract: Communication by speaking is usually done effortlessly, with no thought put into how to produce words or how to decode the sounds we hear into speech. However, speech produces complex sound waves and figuring out which speech sounds these correspond to from the waveforms is not a simple task. In this talk, I will first examine how sound is produced by the larynx. The source-filter theory of speech sound production will then be discussed, along with the anatomy of the vocal tract. The differences between vowels and consonants will be examined. Vowels correspond to specific formants, or amplified harmonics in a spectrogram of the speech, and are thus relatively easy to identify, so the correlations between the articulation of different vowels and the corresponding spectrograms will be examined.

Catherine Felce

Title: The Unreasonable Effectiveness of Mathematics in the Natural Sciences

Abstract: When ‘Nice but Dodgy’ derivations seem to dominate the first and second year physics syllabus, we are naturally left wondering whether there’s more luck than inevitability in the successful discoveries of our subject. Eugene Wigner reflects on a more fundamental disparity between what physicists can know, and what we might expect to be able to. I will follow his argument and consider various responses and philosophical interpretations.