istockphoto-1212503012-612x612.jpg

Wolfson Family Society Easter Egg Painting and Egg Hunt

-
Add to Calendar Wolfson Family Society Easter Egg Painting and Egg HuntThe Buttery
Location
The Buttery
Speakers
N/A
Event price
N/A
Booking Required
Recommended
Please join us for a fun afternoon of Easter egg activities at the Buttery from 2 - 4 pm. We will provide light snacks. Eggs and painting supplies will also be provided.



See you!

Nobuo small.jpg

Nobuo Okaka - Closing Exhibition Event

-
Add to Calendar Nobuo Okaka - Closing Exhibition EventThe Florey Room
Location
The Florey Room
Event price
Free
Booking Required
Not Required
Accessibility
There is provision for wheelchair users.
Meet the artist and join us for a final viewing of Nobuo Okawa's exhibition, From Dark to Light, An Exhibition of Mezzotint Prints on Friday 15 March 2024 from 2-6pm.



Tea and coffee will be available. A last chance to view and purchase a work from this wonderful exhibition.
1549397206672.jpeg

Resurrecting Recurrent Neural Networks for Language Modelling

-
Add to Calendar Resurrecting Recurrent Neural Networks for Language ModellingThe Buttery
Location
The Buttery
Speakers
Dr. Razvan Pascanu (Google DeepMind)
Booking Required
Not Required
Accessibility
There is provision for wheelchair users.
Bio:

I'm currently a Research Scientist at DeepMind. I grew up in Romania and studied computer science and electrical engineering for my undergrads in Germany. I got my MSc from Jacobs University, Bremen in 2009. I hold a PhD from University of Montreal (2014), which I did under the supervision of prof. Yoshua Bengio. I was involved in developing Theano and helped writing some of the deep learning tutorials for Theano. I've published several papers on topics surrounding deep learning and deep reinforcement learning (see my scholar page). I'm one of the organizers of EEML (www.eeml.eu) and part of the organizers of AIRomania. As part of the AIRomania community, I have organized RomanianAIDays since 2020, and helped build a course on AI aimed at high school students.



Abstract:

In this talk I will focus on State Space Models (SSMs) , a subclass of Recurrent Neural Networks (RNNs) that has recently gained some attention through works like Mamba, obtaining strong performance against transformer baselines. I will start by first explaining how SSMs can be viewed as just a particular parametrization of RNNs and what are the crucial differences compared to previous recurrent architectures that led to these results. My goal is to demystify the relative complex parametrization of the architecture and identify what elements are needed for the model to perform well. In this process I will introduce the Linear Recurrent Unit (LRU), a simplified linear layer inspired by existing SSM layers. In the second part of the talk, I will focus on language modelling and the block structure in which such layers tend to be embedded. I will argue that beyond the recurrent layer itself, the block structure borrowed from transformers plays a crucial role in the recent successes of this architecture, and present results at scale of well performing hybrid recurrent architectures as compared to strong transformer baseline. I will close the talk with a few open questions and thoughts on the importance of recurrence in modern deep learning models.

Nikolay Sarkisyan

Marie Curie Post Doctoral Fellow
nikolay.sarkisyan@mod-langs.ox.ac.uk

I am currently a Marie Curie Postdoctoral Research Fellow, delving into a project on historical revolutionary museums in Petrograd-Leningrad during the pivotal years of 1917-1941. This project, funded under the Horizon 2020 European Commission Grant Agreement number 10102852, allows me to explore the intricate transformations of Russian revolutionary traditions in museum spaces, particularly in the course of the 1920s and 1930s. My academic journey commenced at the University of Oslo, where I completed my PhD with a focus on the influence of the discourse of tolerance on contemporary Russian governance. This work offered profound insights into the intersection of politics, society, and history in post-Soviet Russia. My time at the University of Oslo was not only about academic growth but also about enhancing my teaching skills. I contributed to the university's curriculum by teaching courses on contemporary Russian politics and conducting Russian language classes. In 2014-17, before embarking on my doctoral studies, I worked at two prominent (formerly) historical-revolutionary museums: the Museum of Political History of Russia, formerly known as the Museum of the Revolution, and the Smol’ny Museum, previously the Museum of Lenin in Leningrad. My experiences in these museums were not just professionally enriching but also served as a source of inspiration for my current research project. I hold a Master’s degree in Sociology from the European University at St Petersburg and a Specialist degree in History from St Petersburg State University. My research interests are diverse and ever-evolving. Having initially focused on political science for my thesis, I transitioned (back) to history, examining the roots and evolution of Russia's revolutionary tradition. This research is not just a historical inquiry but also engages with broader theoretical questions about the nature of Soviet power, Stalinism, and the 1917 Revolution. In Michaelmas Term 2023, I explored new academic territories, teaching a course on post-Soviet red-brown literature. This endeavor reflects my growing interest in comparative literature and the sociology of literary work, marking a potentially new phase in my academic pursuits.

jakob-foerster.png

Opponent-Shaping and Interference in General-Sum Games

-
Add to Calendar Opponent-Shaping and Interference in General-Sum GamesThe Levett Room
Location
The Levett Room
Speakers
Jakob Foerster
Booking Required
Not Required
Accessibility
There is provision for wheelchair users.
Bio:

Jakob Foerster started as an Associate Professor at the department of engineering science at the University of Oxford in the fall of 2021. During his PhD at Oxford he helped bring deep multi-agent reinforcement learning to the forefront of AI research and interned at Google Brain, OpenAI, and DeepMind. After his PhD he worked as a research scientist at Facebook AI Research in California, where he continued doing foundational work. He was the lead organizer of the first Emergent Communication workshop at NeurIPS in 2017, which he has helped organize ever since and was awarded a prestigious CIFAR AI chair in 2019.





Abstract:

In general-sum games, the interaction of self-interested learning agents commonly leads to collectively worst-case outcomes, such as defect-defect in the iterated prisoner's dilemma (IPD). To overcome this, some methods, such as Learning with Opponent-Learning Awareness (LOLA), shape their opponents' learning process. However, these methods are myopic since only a small number of steps can be anticipated, are asymmetric since they treat other agents as naive learners, and require the use of higher-order derivatives, which are calculated through white-box access to an opponent's differentiable learning algorithm. In this talk I will first introduce Model-Free Opponent Shaping (M-FOS), which overcomes all of these limitations. M-FOS learns in a meta-game in which each meta-step is an episode of the underlying (``inner'') game. The meta-state consists of the inner policies, and the meta-policy produces a new inner policy to be used in the next episode. M-FOS then uses generic model-free optimisation methods to learn meta-policies that accomplish long-horizon opponent shaping. I will finish off the talk with our recent results for adversarial (or cooperative) cheap-talk: How can agents interfere with (or support) the learning process of other agents without being able to act in the environment?
Walker.jpg

MBB Research Cluster: Representation of Sound in the Brain

-
Add to Calendar MBB Research Cluster: Representation of Sound in the BrainSeminar Room 3 - The Academic Wing
Location
Seminar Room 3 - The Academic Wing
Speakers
Associate Professor Kerry Walker
Booking Required
Not Required

The Mind, Brain and Behaviour Research Cluster is pleased to welcome Associate Professor Kerry Walker from the Department of Physiology, Anatomy & Genetics, who will talk about the "representation of sound in the brain", followed by a roundtable discussion.

This event will be held in Seminar Room 3 and is open to all members of the college.

Further information: https://www.wolfson.ox.ac.uk/mind-brain-and-behaviour-mbb-research-cluster

Wolfson College Boatman Completes Record-Breaking Atlantic Crossing

Submitted by george.mather on

Jim Ronaldson has served the Wolfson College Boat Club as a talented and loyal boatman for more than 30 years. Inspired to act by the devastating impact of the pandemic on young mental health, Jim teamed up with fellow rower Ian Davies to take on the "world’s toughest row" to raise funds for Papyrus, which works to prevent suicide amongst young people. Ian raised funds for Myeloma UK.

wolfson pic.jpeg

Wolfson Family Society International Party

-
Add to Calendar Wolfson Family Society International PartyThe Buttery
Location
The Buttery
Speakers
N/A
Event price
N/A
Booking Required
Recommended
The Wolfson Family Society invites all families to our International Party on 2nd March 2024 from 4-6pm at the Buttery. Please bring along a dish from your home country to share (home made or store bought). Please include a note of ingredients in case of allergens. Drinks will be provided. Sign up to help us plan.



See you!



Emilia & Harriet

Jinzhao Sun

jinzhao.sun.phys@gmail.com

Jinzhao Sun is a Visiting Scholar at Wolfson College. Jinzhao received his PhD in physics from the University of Oxford (Jesus College), supervised by Prof Vlatko Vedral and Prof Andrew Boothroyd. He did postdoctoral research with Prof Tom Gur at the University of Cambridge and Prof Myungshik Kim at Imperial College, and he will come back to Oxford supported by a Fellowship. Jinzhao’s research revolves around quantum computing, quantum information, quantum many-body physics, and quantum materials. His research lies in the exploration of quantum computing as a means to address problems that present challenges for classical computing. His research aims to uncover the potential power of quantum computing, and examine whether it holds advantages in tackling quantum many-body problems.

Quantum computing

2023孙金钊 帝国理工.jpg