DCSA

Darwin College Students' Association

Research Talks

All members of Darwin are encouraged to present their research at informal seminars held on Tuesdays and Thursdays during term. Everyone is welcome, whatever your degree or discipline.

Darwin members pick up lunch from 12:00, taking it into the Richard King Room (on the left at the top of the stairs leading to the dining hall) or 1 Newnham Terrace (straight through at the far end of the dining hall). Wine is served. Non-Darwin members are welcome to attend, although lunch is only available to guests of members. The talk begins at about 1:15 and lasts for about 20 minutes and is followed by questions over coffee. We adjourn at 2:00pm at the latest.

Upcoming Talks

Thursday 23 November 2017
The Richard King Room, Darwin College
KC Sivaramakrishnan (Computer Laboratory, University of Cambridge)

Cryptocurrencies such as Bitcoin and Ethereum have gained enormous traction in the last few years. The monumental increase in value of Bitcoin and other cryptocurrencies seems only to be topped by the hype and notoriety surrounding them. But at its core, the Blockchain -- the technology at the foundation of cryptocurrencies -- has very desirable properties such as integrity, resilience and transparency, which makes it appealing to a large class of decentralized applications. In this talk, I will discuss some of the foundational technology that underlies the Blockchain. I will follow this up with some of the open challenges in this space and my current research threads that overlap with them.

Tuesday 28 November 2017
The Richard King Room, Darwin College
Thomas McGuire, Research Fellow, Darwin College

One of the several research projects Tom is currently pursuing is assessing the influence of the UK on the development of state security sectors in the Global South - in particular but not exclusively the Commonwealth - through training, equipment and other forms of assistance since 1945. This is intended to better inform understanding of, on the one hand, the UK’s post-colonial legacies and foreign policy and, on the other, contemporary debates regarding upstream conflict prevention, human rights, and security sector reform why security sectors develop in similar and different ways. This talk will present preliminary findings from one case area of the project: Cold War Southeast Asia, placing it in the context of British overseas security assistance and foreign policy across the Global South in this era.

Thursday 30 November 2017
The Richard King Room, Darwin College
Katarina Pisani (Department of Chemistry, University of Cambridge)

In the last decades, there has been a rapid demographic shift, where populations in both developing and developed countries live far longer. Although an indication of medical advances and overall improved health, an increase in lifespan comes with great costs too. Individuals over the age of 65 have an increased chance of developing dementias and neurodegenerative diseases, such as Alzheimer’s and Parkinson’s disease, and the chances increase every year. Despite numerous clinical trials and funds invested in testing for new cures and treatments, nothing has yet been found. These diseases, which are still incurable, progressive and eventually fatal, currently represent a tremendous burden on our social systems, as well as the patients’ and their families’ lives. The primary reason why no significant development in treating these conditions has occurred is that we do not really understand their molecular origins. In the Centre for Misfolding diseases we have been working to develop a ‘gene signature’ for such conditions, which will provide us with a tool to gain insight and allow us to recapitulate these diseases, which will test our fundamental understanding of their causes, as well as enabling effective drug discovery programs to be carried out.

Past Research Talks

Thursday 12 October 2017
Giancarlo Soavi (Cambridge Graphene Centre)

Laser sources producing nanosecond (10-9 s) to sub-picosecond (10-12 s) pulses (i.e. ultrafast lasers) are deployed in a variety of applications ranging from scientific research, laser surgery, material processing and telecommunications. Regardless of the output wavelength, the majority of ultrafast laser systems employ a mode-locking technique, whereby a nonlinear optical element - called Saturable Absorber (SA) - turns the laser continuous wave output into a train of ultrashort optical pulses. The SA absorption (or optical loss) decreases as the incident light intensity increases. Thus, the SA works as an intensity-dependent optical switch. The key requirements for SAs are fast response time, high modulation depth, broad wavelength range, low optical loss, low-cost and ease of integration into an optical system. Graphene, a one atom thick layer of carbon atoms arranged in a honeycomb lattice, can simultaneously meet all these needs with better performances and lower cost compared to current technologies. In this seminar I will introduce the basic concepts of ultrafast lasers and mode-locking and their importance for technological applications. I will then review the fundamental physical properties that make graphene the ideal candidate as saturable absorber for ultrafast lasers on an extremely broad energy range from visible to THz.

Tuesday 10 October 2017
Daniel H. Weiss, Polonsky-Coexist Senior Lecturer in Jewish Studies in the Faculty of Divinity

Many scholars today view the causing the death of innocent civilians in warfare as an established part of historical Western tradition of just war, so long as those deaths are 'merely foreseen, but not directly intended'. This attitude towards 'collateral damage' or 'double effect' is often traced back to Thomas Aquinas. However, I argue that, contrary to received scholarly assumptions, Aquinas in fact sharply rejects the legitimacy of such forms of killing. Accordingly, premodern Western thought regarding just war may stand in a much sharper discontinuity with modern just war ethics than has previously been recognized, with significant potential implications for contemporary public debates and ethical dilemmas.

Thursday 5 October 2017
Dr. Peter Murray-Rust (Unilever Centre, Dept. Of Chemistry, University of Cambridge)

Peter Murray-Rust, ContentMine [1] and University of Cambridge

Public funding of science and medicine generates 1 trillion dollars of public knowledge per year but most of this is inaccessible to most people. Working with the Wikimedia Foundation we have developed tools for collecting over 6 million of the world's open scientific articles and extracting the facts from them into WikiFactMine (WFM) [2] . We use Wikidata [3] which, with over 40 million "items" from Wikipedia or world authorities, is based on modern Open Web technology. WFM reads every new Open scientific article (starting with biomedicine) and indexes the terms against WikiFactMine. It thus becomes a "knowledge prosthetic" or "amanuensis" so that everyone can immediately find the accumulated knowledge in Wikimedia resources.

We believe that with WikiFactMine the scientific literature becomes accessible to a wide range of people and machines. Data in articles can be automatically indexed on fulltext and diagrammatic content creating the base for a new generation of scientific search engines. We have created a wide range of "dictionaries" from Wikidata, allowing multidisciplinary search of articles (e.g. chemistry, diseases, drugs...) . WikiFactMine can expand "find all chemicals produced by conifers" to 500 phytochemicals and 2000 conifers and search for all of them. "What viral diseases have been reported in West Africa" might inform public health policies in a new manner.

The talk will cover the technology (which anyone can use; ContentMine already has a 15-year old contributing) and the politics of academic publication where revenue is often generated by artificial scarcity. Can we find a better way? Everyone can participate in WikiFactMine.

I thank Charles Matthews and Tom Arrow who created WikiFactMine.

[1] http://contentmine.org [2] https://www.wikidata.org/wiki/Wikidata:WikiFactMine [3] https://www.wikidata.org

Thursday 25 May 2017
Shehar Bano, University College London

Censorship of online communications threatens principles of openness and freedom of information on which the Internet was founded. In the interest of transparency and accountability, and more broadly to develop scientific rigour in the field, we need methodologies to measure and characterize Internet censorship. Such studies will not only help users make informed choices about information access, but also illuminate entities involved in or affected by censorship; informing the development of policy and enquiries into the ethics and legality of such practices. However, many issues around Internet censorship remain poorly understood because of the inherently adversarial and opaque landscape in which it operates. As details about mechanisms and targets of censorship are usually undisclosed, it is hard to define exactly what comprises censorship, and how it operates in different contexts.

My research aims to help fill this gap by developing methodologies to derive censorship ground truth using active and passive data analysis techniques, which I apply to real-world datasets to uncover entities involved in censorship, the targets of censorship, and the effects of such practices on different stakeholders. In this talk, I will provide an overview of my work on Internet censorship from multiple perspectives: (i) measurement of the Great Firewall of China that shows that inference of the censor’s traffic analysis model can enable systematic identification of evasion opportunities that users can exploit to access restricted content, (ii) analysis of network logs collected at an Internet Service Provider (ISP) in Pakistan over a period of escalating censorship to study how censorship affects users’ browsing habits with respect to circumvention, and its economic effects on content providers and ISPs, and (iii) investigation of differential treatment -- an emerging class of censorship where websites (rather than the government) block requests of users they don’t like -- in the context of Tor anonymity network and users of adblocking software.

Tuesday 23 May 2017
Pablo Salas (University of Cambridge)

Collaboration between researchers and policy-makers has perhaps never been as crucial as it is today, in view of the many critical issues that countries, particularly Brazil, face in the context of the Water-Energy-Food (FEW) nexus. A perfect storm of complex interactions, dependencies and vulnerabilities is most likely to be expected in Brazil, given its current environmental and economic situation. On the one hand, climate change is highly likely to change weather patterns, which will detrimentally affect agriculture and biodiversity in Brazil. On the other hand, Brazilian economy relies heavily on exports of natural resources for prosperity, and global changes in demand for commodities will put pressure on the Brazilian economy. In this talk, I will present the main aspects of the complex nexus system, with special focus on the challenges associated to create policy to improve the resilience of the Brazilian FEW Nexus.

Dr Pablo Salas is an Economist and Electrical Engineer by training, with a PhD in Land Economy from the University of Cambridge. He is currently a Research Fellow at the Cambridge Centre for Environment, Energy and Natural Resource Governance (C-EENRG), Department of Land Economy. Dr Salas' wider research examines how interactions among energy, environmental and economic systems can be used to improve global strategies for climate change risk reduction and sustainable economic development. As part of his fellowship, he is also leading the development of various outreach activities at C-EENRG, actively connecting academics with policy makers and innovators.

Close menu
Site navigation mobile menu