Spring 2019

CNU lab meetings are usually at 3pm on Fridays. See also our new list highlighting CNU seminars and other computational neuroscience seminars across the university. More info on bristolcnu.github.io


1 February – SNOW!


8  February – Marie Tolkiehn 

“Neural ensemble activity depends on stimulus type in mouse primary visual cortex”

2.59 MVB
Already in the 1950s it was discovered that moving bars at different orientations elicited responses of varying strength in primary visual cortex. This led to the discovery of orientation-selective neurons (Hubel and Wiesel 1959) and much research into edge detection. However, neural ensemble activity differs between stimulus types: Artificial stimuli such as moving bars or gratings induce responses different from spontaneous activity (no stimulation) or evoked by natural scenes. We found spontaneous and natural scenes-evoked activities indicated lower firing rates, Shannon entropies, and binary word distribution divergences (Jensen-Shannon-Divergence) than either to drifting gratings.

15 February – Cian

journal club talk on

Fundamental bounds on learning performance in neural circuits

Dhruva V Raman, Timothy O’Leary

https://www.biorxiv.org/content/10.1101/508994v1


22 February – Conor

journal club talk on

Mikolov, Tomas, et al. “Efficient estimation of word representations in vector space.arXiv preprint arXiv:1301.3781 (2013).

https://arxiv.org/pdf/1301.3781.pdf


1 March – Alastair Craw (Bath)

A novel prism-glasses based treatment for CFS.


Date TBA – Joint meeting with Collective Dynamics


15 March – Jack Mellor (PPN)

A new model for neuromodulation: Integrated encoding of expectation.

Neuromodulator systems provide the principle signals to achieve rapid behaviourally-relevant adaptations in brain function. However, our understanding of their mechanism of action is hampered by current models of the different neuromodulator systems as segregated signalling modalities. In this talk I will propose a new model for neuromodulation where distinct but functionally overlapping neuromodulator systems combine and interact to signal different aspects of predictions, prediction errors and uncertainty, i.e. expectation. I will describe how we might test these models by measuring where and when neuromodulation by dopamine, serotonin, noradrenaline and acetylcholine occurs in prefrontal cortex during a behavioural task in mice designed to manipulate the type and degree of uncertainty.


22 March – Milton Montero (Journal club)

https://www.nature.com/articles/s41593-018-0310-2

Task representations in neural networks trained to perform many cognitive tasks, Guangyu Robert Yang, Madhura R. Joglekar, H. Francis Song, William T. Newsome & Xiao-Jing Wan


29 March – no meeting


5 April – Jeff Mitchell (EP)

Extrapolation in Deep-Learning

Historically, criticism of neural or connectionist models of cognition has often used symbolic architectures as a contrast to highlight shortcomings, with systematicity and productivity being two characteristics that are frequently asserted to be lacking in the former but not the latter. Recently, Marcus re-opened this dialogue with a  widely discussed essay that criticised deep learning from a symbolic perspective on a number of fronts. This talk will focus on the issue of extrapolation and Marcus’s identity learning challenge, dating back to 1998.

While the ability to successfully make novel predictions outside the original domain is usually considered a critical test of scientific models and theories, the standard machine learning methodology focuses almost exclusively on within domain testing. Extrapolating outside the training data may require different approaches, in particular architectures that connect distant regions of the input space. I will discuss the role of symmetry in this regard and present a couple of examples.


12 April -. Orlando da Silva.

Orlando is a consultant ophthalmologist with interest in proprioceptive disorders and their treatment through optical means. More about his background can be found on this site

https://www.orlandoalvesdasilva.org/

The work has links with neuroscience, psychology and medicine; for example, relating to the recent talk by Dr Alastair Craw, mathematician from the University of Bath, on his experiences with M.E.


19 April – Good Friday


26 April – Roberto Feuda (Palaeobiology and Biodiversity Research Group)

Towards a system-level understanding of the evolution of the nervous system


3 May – no talk


10 May – Anne-Lene Sax (SCEEM/EP, Bristol)

Towards understanding of self-beliefs in the context of depression


17 May – Interviews for new post


24 May – Sean-James Fallon (Bristol) 0.2 MVB

The effect of dopamine and Parkinson’s disease on the control of irrelevant information.


7 June – Thom Griffith / Sophie Baker (Bristol) 5.68 Wills


11 June – CNU seminar: Irene Malvestio (Univ. Pompeu Fabra, Spain)

Venue: 3.33 Wills Memorial Building      Time: 4pm

Detection of directional interactions between single neurons: comparison of methods and application to sleep recordings

11 June – CNU seminar: Marinho Lopes (University of Bristol)

Venue: 3.33 Wills Memorial Building      Time: 11:30am

Mathematical modelling of epilepsy and its applications in epilepsy surgery and epilepsy classification

14 June – Moved to later in the year

Simon Bright (PPN, Bristol)

On Deeplabcut: http://www.mousemotorlab.org/deeplabcut


21 June – CNU Seminar: Katharina Wilmes (Imperial College London) 4.01 MVB

Inhibitory microcircuits for top-down plasticity of sensory representations


28 June – Paul Dodson (PPN, Bristol) 4.01 MVB

A Mötley Crüe: Heterogeneity in the encoding of behaviour by dopamine neurons


5 July – Daniel Bennett (SCEEM, Bristol) – TBC 4.01 MVB


July – September – summer hiatus


 

Venue schedule:

2.59 MVB: 1st Feb / 8th Feb / 15th Feb / 22nd Feb / 1st March / 8th March

4.01 MVB: 15th March / 22nd March / 29th March

2.59 MVB: 5th April / 12th April / 19th April / 26th April / 3rd May

0.3 MVB: 10th May

 

Leave a Reply

Your email address will not be published. Required fields are marked *