Amos Storkey
@AmosStorkey
Followers
391
Following
15
Media
1
Statuses
33
Joined December 2014
One or two deep learning postdocs in Edinburgh with Amos Storkey: continual few-shot learning and beyond - https://t.co/wewtQFcSmA
#AI #MachineLearning #MLjobs #DataScience #AIjobs
0
19
18
I am advertising a PhD studentship in Better Sample Efficiency in Deep Reinforcement Learning, joint with Kamil Ciosek and Microsoft Research Cambridge. https://t.co/azNpU9GbVj
0
0
4
Meta-learning and AutoML folks. Are you tired of being limited to a few inner steps when learning hyperparameters with gradients? Our latest paper enables gradient-based HPO over CIFAR-10 sized problems, without introducing greediness. https://t.co/zX7kt4DL52
0
10
31
Lecturer/Reader (equiv. Assist./Assoc. Prof.) in Machine Learning. School of Informatics, University of Edinburgh. Deadline: 5pm 22 Jan 2020. #mljobs #AIjobs #facultyjobs #academicjobs #jobsacuk #DeepLearning
https://t.co/M7tUYXfoRd
0
15
12
Lecturer/Reader (equiv. Assist./Assoc. Prof.) in Machine Learning. School of Informatics, University of Edinburgh. Deadline: 5pm 22 Jan 2020. https://t.co/M7tUYXfoRd
0
2
2
Our work on Gaussian Processes and Deep Neural Networks for few-shot learning.
0
2
8
First paper, first tweet! Here we consider an increasingly common problem: model compression when training data is not available (e.g. private or too large). Our solution is to adversarially search for pseudo data, and use it for distillation. Paper & code
arxiv.org
Performing knowledge transfer from a large teacher network to a smaller student is a popular task in modern deep learning applications. However, due to growing dataset sizes and stricter privacy...
Zero-shot Knowledge Transfer via Adversarial Belief Matching
2
8
17
Humans can learn new notions quickly from a small number of labelled examples. However, they can also adapt those notions in light of new unlabelled data, to improve their generalization. Inspired by this, we develop 'Self-Critique and Adapt'.
arxiv.org
In few-shot learning, a machine learning system learns from a small set of labelled examples relating to a specific task, such that it can generalize to new examples of the same task. Given the...
2
11
48
Scott Rudin kills off performances of To Kill a Mockingbird across the world with threats of law suits. Acceptable? Not to me. https://t.co/3MlRiLOWNf
#boycottRudinplays
0
0
0
Edinburgh postdoc position available. Novel methods for Deep Learning in Brain Imaging. https://t.co/4eeiIxtUwE . Joint between Informatics https://t.co/DzMOk0rAEc Dundee https://t.co/H0XmETEeWP and Centre for Clinical Brain Sciences https://t.co/ChNykqAJHF . #Bayeswatch @InfEd
0
6
8
My latest blog post on meta-learning in general and "How to train your MAML" in particular is now out. https://t.co/n25TwwQiEZ The post thoroughly explains MAML, some of its problems, and proposes some solutions. In addition visualizes the learned per-step per layer learning rate
1
8
41
Machine Learning faculty posts @Informatics, Edinburgh, UK. Substantial expansion in ML&AI, New Bayes Centre for Data Science. Speak to us (Amos Storkey, Chris Williams) at NeurIPS. https://t.co/MxtsFgZlIw
#NIPS #NeurIPS #Edinburgh #DeepLearning #mljobs @InfAtEd
elxw.fa.em3.oraclecloud.com
Help make the World a better place
0
9
20
It is the early part of Neural Network training that matters: Fisher pruning appears ineffective without retraining.
Pruning neural networks: is it time to nip it in the bud?. Elliot J. Crowley, Jack Turner, Amos Storkey, and Michael O'Boyle
0
0
3
CINIC-10: an extended plug in for CIFAR-10 Report: https://t.co/AfqnhehXGf Blog: https://t.co/1rBCr2UKax Data: https://t.co/0VRkgOiXWg Github:
github.com
A drop-in replacement for CIFAR-10. Contribute to BayesWatch/cinic-10 development by creating an account on GitHub.
0
0
6
What do neural network functions look like as they learn? GINN: Geometric Illustrations for Neural Networks. https://t.co/oJ1LKgSPSc and https://t.co/kBS2mJQcec
#bayeswatch
0
4
6
Postdoc -- Informatics, Edinburgh: Deep Learning Methods for Constrained Environments. Start Date Flexible. Deadline to be extended 1 week to 25 May.
0
16
11
Interested in PhD place in Data Science, Machine Learning, AI?
0
5
3
State of the art one shot learning via a Data Augmentation GAN: DAGAN https://t.co/lKx4J100hp
arxiv.org
Effective training of neural networks requires much data. In the low-data regime, parameters are underdetermined, and learnt networks generalise poorly. Data Augmentation...
0
1
9
Deep Learning postdoc, Informatics, Edinburgh. Closing soon (Mar 31).
0
0
1