Search Results for author: Timothy Hospedales

Found 73 papers, 23 papers with code

Zero-Shot Everything Sketch-Based Image Retrieval, and in Explainable Style

1 code implementation25 Mar 2023 Fengyin Lin, Mingkang Li, Da Li, Timothy Hospedales, Yi-Zhe Song, Yonggang Qi

This paper studies the problem of zero-short sketch-based image retrieval (ZS-SBIR), however with two significant differentiators to prior art (i) we tackle all variants (inter-category, intra-category, and cross datasets) of ZS-SBIR with just one network (``everything''), and (ii) we would really like to understand how this sketch-photo matching operates (``explainable'').

Retrieval Sketch-Based Image Retrieval

Amortised Invariance Learning for Contrastive Self-Supervision

no code implementations24 Feb 2023 Ruchika Chavhan, Henry Gouk, Jan Stuehmer, Calum Heggan, Mehrdad Yaghoobi, Timothy Hospedales

Contrastive self-supervised learning methods famously produce high quality transferable representations by learning invariances to different data augmentations.

Contrastive Learning Representation Learning +1

Domain Generalisation via Domain Adaptation: An Adversarial Fourier Amplitude Approach

no code implementations23 Feb 2023 Minyoung Kim, Da Li, Timothy Hospedales

We tackle the domain generalisation (DG) problem by posing it as a domain adaptation (DA) task where we adversarially synthesise the worst-case target domain and adapt a model to that worst-case domain, thereby improving the model's robustness.

Domain Adaptation

Meta-Learned Kernel For Blind Super-Resolution Kernel Estimation

no code implementations15 Dec 2022 Royson Lee, Rui Li, Stylianos I. Venieris, Timothy Hospedales, Ferenc Huszár, Nicholas D. Lane

Recent image degradation estimation methods have enabled single-image super-resolution (SR) approaches to better upsample real-world images.

Blind Super-Resolution Image Super-Resolution

Federated Learning for Inference at Anytime and Anywhere

no code implementations8 Dec 2022 Zicheng Liu, Da Li, Javier Fernandez-Marques, Stefanos Laskaridis, Yan Gao, Łukasz Dudziak, Stan Z. Li, Shell Xu Hu, Timothy Hospedales

Federated learning has been predominantly concerned with collaborative training of deep networks from scratch, and especially the many challenges that arise, such as communication cost, robustness to heterogeneous data, and support for diverse device capabilities.

Federated Learning

MEDFAIR: Benchmarking Fairness for Medical Imaging

1 code implementation4 Oct 2022 Yongshuo Zong, Yongxin Yang, Timothy Hospedales

In this work, we introduce MEDFAIR, a framework to benchmark the fairness of machine learning models for medical imaging.

Benchmarking Fairness +2

Attacking Adversarial Defences by Smoothing the Loss Landscape

1 code implementation1 Aug 2022 Panagiotis Eustratiadis, Henry Gouk, Da Li, Timothy Hospedales

This paper investigates a family of methods for defending against adversarial attacks that owe part of their success to creating a noisy, discontinuous, or otherwise rugged loss landscape that adversaries find difficult to navigate.


HyperInvariances: Amortizing Invariance Learning

no code implementations17 Jul 2022 Ruchika Chavhan, Henry Gouk, Jan Stühmer, Timothy Hospedales

Providing invariances in a given learning task conveys a key inductive bias that can lead to sample-efficient learning and good generalisation, if correctly specified.

Inductive Bias

Feed-Forward Source-Free Latent Domain Adaptation via Cross-Attention

no code implementations15 Jul 2022 Ondrej Bohdal, Da Li, Shell Xu Hu, Timothy Hospedales

We study the highly practical but comparatively under-studied problem of latent-domain adaptation, where a source model should be adapted to a target dataset that contains a mixture of unlabelled domain-relevant and domain-irrelevant examples.

Source-Free Domain Adaptation

MetaAudio: A Few-Shot Audio Classification Benchmark

1 code implementation5 Apr 2022 Calum Heggan, Sam Budgett, Timothy Hospedales, Mehrdad Yaghoobi

Currently available benchmarks for few-shot learning (machine learning with few training examples) are limited in the domains they cover, primarily focusing on image classification.

Audio Classification Classification +2

Finding lost DG: Explaining domain generalization via model complexity

no code implementations1 Feb 2022 Da Li, Henry Gouk, Timothy Hospedales

However much of the work in general purpose DG is heuristically motivated, as the DG problem is hard to model formally; and recent evaluations have cast doubt on existing methods' practical efficacy -- in particular compared to a well tuned empirical risk minimisation baseline.

Domain Generalization

Gaussian Process Meta Few-shot Classifier Learning via Linear Discriminant Laplace Approximation

no code implementations9 Nov 2021 Minyoung Kim, Timothy Hospedales

In essence, the MAP solution is approximated by the LDA estimate, but to take the GP prior into account, we adopt the prior-norm adjustment to estimate LDA's shared variance parameters, which ensures that the adjusted estimate is consistent with the GP prior.


Defensive Tensorization

no code implementations26 Oct 2021 Adrian Bulat, Jean Kossaifi, Sourav Bhattacharya, Yannis Panagakis, Timothy Hospedales, Georgios Tzimiropoulos, Nicholas D Lane, Maja Pantic

We propose defensive tensorization, an adversarial defence technique that leverages a latent high-order factorization of the network.

Audio Classification Classification +1

Online Hyperparameter Meta-Learning with Hypergradient Distillation

no code implementations ICLR 2022 Hae Beom Lee, Hayeon Lee, Jaewoong Shin, Eunho Yang, Timothy Hospedales, Sung Ju Hwang

Many gradient-based meta-learning methods assume a set of parameters that do not participate in inner-optimization, which can be considered as hyperparameters.

Hyperparameter Optimization Knowledge Distillation +1

SketchODE: Learning neural sketch representation in continuous time

no code implementations ICLR 2022 Ayan Das, Yongxin Yang, Timothy Hospedales, Tao Xiang, Yi-Zhe Song

Learning meaningful representations for chirographic drawing data such as sketches, handwriting, and flowcharts is a gateway for understanding and emulating human creative expression.

Data Augmentation

Loss Function Learning for Domain Generalization by Implicit Gradient

no code implementations29 Sep 2021 Boyan Gao, Henry Gouk, Yongxin Yang, Timothy Hospedales

We take a different approach, and explore the impact of the ERM loss function on out-of-domain generalisation.

Domain Generalization Meta-Learning

Vision-based system identification and 3D keypoint discovery using dynamics constraints

no code implementations13 Sep 2021 Miguel Jaques, Martin Asenov, Michael Burke, Timothy Hospedales

This paper introduces V-SysId, a novel method that enables simultaneous keypoint discovery, 3D system identification, and extrinsic camera calibration from an unlabeled video taken from a static camera, using only the family of equations of motion of the object of interest as weak supervision.

Camera Calibration

A Channel Coding Benchmark for Meta-Learning

1 code implementation15 Jul 2021 Rui Li, Ondrej Bohdal, Rajesh Mishra, Hyeji Kim, Da Li, Nicholas Lane, Timothy Hospedales

We use our MetaCC benchmark to study several aspects of meta-learning, including the impact of task distribution breadth and shift, which can be controlled in the coding problem.


EvoGrad: Efficient Gradient-Based Meta-Learning and Hyperparameter Optimization

1 code implementation NeurIPS 2021 Ondrej Bohdal, Yongxin Yang, Timothy Hospedales

Gradient-based meta-learning and hyperparameter optimization have seen significant progress recently, enabling practical end-to-end training of neural networks together with many hyperparameters.

cross-domain few-shot learning Hyperparameter Optimization

Meta-Calibration: Learning of Model Calibration Using Differentiable Expected Calibration Error

1 code implementation17 Jun 2021 Ondrej Bohdal, Yongxin Yang, Timothy Hospedales

The problem is especially noticeable when using modern neural networks, for which there is a significant difference between the confidence of the model and the probability of correct prediction.


Cloud2Curve: Generation and Vectorization of Parametric Sketches

1 code implementation CVPR 2021 Ayan Das, Yongxin Yang, Timothy Hospedales, Tao Xiang, Yi-Zhe Song

Analysis of human sketches in deep learning has advanced immensely through the use of waypoint-sequences rather than raster-graphic representations.

Shallow Bayesian Meta Learning for Real-World Few-Shot Recognition

2 code implementations ICCV 2021 Xueting Zhang, Debin Meng, Henry Gouk, Timothy Hospedales

Current state-of-the-art few-shot learners focus on developing effective training procedures for feature representations, before using simple, e. g. nearest centroid, classifiers.

cross-domain few-shot learning Few-Shot Image Classification

Reducing Implicit Bias in Latent Domain Learning

no code implementations1 Jan 2021 Lucas Deecke, Timothy Hospedales, Hakan Bilen

A fundamental shortcoming of deep neural networks is their specialization to a single task and domain.

Image Classification

Robust Domain Randomised Reinforcement Learning through Peer-to-Peer Distillation

no code implementations9 Dec 2020 Chenyang Zhao, Timothy Hospedales

In reinforcement learning, domain randomisation is an increasingly popular technique for learning more general policies that are robust to domain-shifts at deployment.

Continuous Control reinforcement-learning +1

Margin-Based Transfer Bounds for Meta Learning with Deep Feature Embedding

no code implementations2 Dec 2020 Jiechao Guan, Zhiwu Lu, Tao Xiang, Timothy Hospedales

By transferring knowledge learned from seen/previous tasks, meta learning aims to generalize well to unseen/future tasks.

Classification General Classification +2

Weight-Covariance Alignment for Adversarially Robust Neural Networks

no code implementations17 Oct 2020 Panagiotis Eustratiadis, Henry Gouk, Da Li, Timothy Hospedales

Stochastic Neural Networks (SNNs) that inject noise into their hidden layers have recently been shown to achieve strong robustness against adversarial attacks.

Adversarial Robustness

Learning the Prediction Distribution for Semi-Supervised Learning with Normalising Flows

1 code implementation6 Jul 2020 Ivana Balažević, Carl Allen, Timothy Hospedales

In this work, we propose a probabilistically principled general approach to SSL that considers the distribution over label predictions, for labels of different complexity, from "one-hot" vectors to binary vectors and images.

General Classification Image Classification +3

BézierSketch: A generative model for scalable vector sketches

1 code implementation ECCV 2020 Ayan Das, Yongxin Yang, Timothy Hospedales, Tao Xiang, Yi-Zhe Song

The study of neural generative models of human sketches is a fascinating contemporary modeling problem due to the links between sketch image generation and the human drawing process.

Image Generation

Flexible Dataset Distillation: Learn Labels Instead of Images

2 code implementations15 Jun 2020 Ondrej Bohdal, Yongxin Yang, Timothy Hospedales

In particular, we study the problem of label distillation - creating synthetic labels for a small set of real images, and show it to be more effective than the prior image-based approach to dataset distillation.

Data Summarization Meta-Learning

A Probabilistic Model for Discriminative and Neuro-Symbolic Semi-Supervised Learning

no code implementations10 Jun 2020 Carl Allen, Ivana Balažević, Timothy Hospedales

Much progress has been made in semi-supervised learning (SSL) by combining methods that exploit different aspects of the data distribution, e. g. consistency regularisation relies on properties of $p(x)$, whereas entropy minimisation pertains to the label distribution $p(y|x)$.

Image Augmentation Logical Reasoning

NewtonianVAE: Proportional Control and Goal Identification from Pixels via Physical Latent Spaces

no code implementations CVPR 2021 Miguel Jaques, Michael Burke, Timothy Hospedales

Learning low-dimensional latent state space dynamics models has been a powerful paradigm for enabling vision-based planning and learning for control.

Behavioural cloning

Latent Domain Learning with Dynamic Residual Adapters

no code implementations1 Jun 2020 Lucas Deecke, Timothy Hospedales, Hakan Bilen

While recent techniques in domain adaptation and multi-domain learning enable the learning of more domain-agnostic features, their success relies on the presence of domain labels, typically requiring manual annotation and careful curation of datasets.

Domain Adaptation Image Classification +1

SimpleMKKM: Simple Multiple Kernel K-means

1 code implementation11 May 2020 Xinwang Liu, En Zhu, Jiyuan Liu, Timothy Hospedales, Yang Wang, Meng Wang

We propose a simple yet effective multiple kernel clustering algorithm, termed simple multiple kernel k-means (SimpleMKKM).

Meta-Learning in Neural Networks: A Survey

1 code implementation11 Apr 2020 Timothy Hospedales, Antreas Antoniou, Paul Micaelli, Amos Storkey

We survey promising applications and successes of meta-learning such as few-shot learning and reinforcement learning.

Few-Shot Learning Hyperparameter Optimization +1

Online Meta-Learning for Multi-Source and Semi-Supervised Domain Adaptation

no code implementations ECCV 2020 Da Li, Timothy Hospedales

Therefore we propose an online shortest-path meta-learning framework that is both computationally tractable and practically effective for improving DA performance.

Meta-Learning Multi-Source Unsupervised Domain Adaptation +1

Sequential Learning for Domain Generalization

no code implementations3 Apr 2020 Da Li, Yongxin Yang, Yi-Zhe Song, Timothy Hospedales

In DG this means encountering a sequence of domains and at each step training to maximise performance on the next domain.

Domain Generalization Meta-Learning

Deep Domain-Adversarial Image Generation for Domain Generalisation

no code implementations12 Mar 2020 Kaiyang Zhou, Yongxin Yang, Timothy Hospedales, Tao Xiang

This is achieved by having a learning objective formulated to ensure that the generated data can be correctly classified by the label classifier while fooling the domain classifier.

Domain Generalization Image Generation

Incremental Few-Shot Object Detection

no code implementations CVPR 2020 Juan-Manuel Perez-Rua, Xiatian Zhu, Timothy Hospedales, Tao Xiang

To this end we propose OpeN-ended Centre nEt (ONCE), a detector designed for incrementally learning to detect novel class objects with few examples.

Few-Shot Learning Few-Shot Object Detection +2

Adversarial Generation of Informative Trajectories for Dynamics System Identification

1 code implementation2 Mar 2020 Marija Jegorova, Joshua Smith, Michael Mistry, Timothy Hospedales

Dynamic System Identification approaches usually heavily rely on the evolutionary and gradient-based optimisation techniques to produce optimal excitation trajectories for determining the physical parameters of robot platforms.

Full-Scale Continuous Synthetic Sonar Data Generation with Markov Conditional Generative Adversarial Networks

no code implementations15 Oct 2019 Marija Jegorova, Antti Ilari Karjalainen, Jose Vazquez, Timothy Hospedales

High-quality realistic sonar data simulation could be of benefit to multiple applications, including training of human operators for post-mission analysis, as well as tuning and validation of autonomous target recognition (ATR) systems for underwater vehicles.

Interpreting Knowledge Graph Relation Representation from Word Embeddings

no code implementations ICLR 2021 Carl Allen, Ivana Balažević, Timothy Hospedales

Many models learn representations of knowledge graph data by exploiting its low-rank latent structure, encoding known relations between entities and enabling unknown facts to be inferred.

Link Prediction Word Embeddings

Simple and Effective Stochastic Neural Networks

no code implementations25 Sep 2019 Tianyuan Yu, Yongxin Yang, Da Li, Timothy Hospedales, Tao Xiang

Stochastic neural networks (SNNs) are currently topical, with several paradigms being actively investigated including dropout, Bayesian neural networks, variational information bottleneck (VIB) and noise regularized learning.

Adversarial Attack Adversarial Defense

Zero-Shot Crowd Behavior Recognition

no code implementations16 Aug 2019 Xun Xu, Shaogang Gong, Timothy Hospedales

To that end, we relax the common assumption that each individual crowd video instance is only associated with a single crowd attribute.

Zero-Shot Learning

Factorized Higher-Order CNNs with an Application to Spatio-Temporal Emotion Estimation

no code implementations CVPR 2020 Jean Kossaifi, Antoine Toisoul, Adrian Bulat, Yannis Panagakis, Timothy Hospedales, Maja Pantic

To alleviate this, one approach is to apply low-rank tensor decompositions to convolution kernels in order to compress the network and reduce its number of parameters.

Emotion Recognition Image Classification

Physics-as-Inverse-Graphics: Unsupervised Physical Parameter Estimation from Video

1 code implementation ICLR 2020 Miguel Jaques, Michael Burke, Timothy Hospedales

Our approach significantly outperforms related unsupervised methods in long-term future frame prediction of systems with interacting objects (such as ball-spring or 3-body gravitational systems), due to its ability to build dynamics into the model as an inductive bias.

Inductive Bias Scene Understanding +1

Multi-relational Poincaré Graph Embeddings

1 code implementation NeurIPS 2019 Ivana Balažević, Carl Allen, Timothy Hospedales

Hyperbolic embeddings have recently gained attention in machine learning due to their ability to represent hierarchical data more accurately and succinctly than their Euclidean analogues.

Entity Embeddings Knowledge Graphs +1

Analogies Explained: Towards Understanding Word Embeddings

no code implementations28 Jan 2019 Carl Allen, Timothy Hospedales

Word embeddings generated by neural network methods such as word2vec (W2V) are well known to exhibit seemingly linear behaviour, e. g. the embeddings of analogy "woman is to queen as man is to king" approximately describe a parallelogram.

Word Embeddings

Behavioural Repertoire via Generative Adversarial Policy Networks

no code implementations7 Nov 2018 Marija Jegorova, Stéphane Doncieux, Timothy Hospedales

Leveraging our generative policy network, a robot can sample novel behaviors until it finds one that works for a new environment.

Learning Unsupervised Word Translations Without Adversaries

no code implementations EMNLP 2018 Tanmoy Mukherjee, Makoto Yamada, Timothy Hospedales

Word translation, or bilingual dictionary induction, is an important capability that impacts many multilingual language processing tasks.

Machine Translation Multilingual Word Embeddings +3

Meta-Learning Transferable Active Learning Policies by Deep Reinforcement Learning

no code implementations ICLR 2018 Kunkun Pang, Mingzhi Dong, Yang Wu, Timothy Hospedales

In contrast to this body of research, we propose to treat active learning algorithm design as a meta-learning problem and learn the best criterion from data.

Active Learning Meta-Learning +2

What the Vec? Towards Probabilistically Grounded Embeddings

no code implementations NeurIPS 2019 Carl Allen, Ivana Balažević, Timothy Hospedales

We show that different interactions between PMI vectors reflect semantic word relationships, such as similarity and paraphrasing, that are encoded in low dimensional word embeddings under a suitable projection, theoretically explaining why embeddings of W2V and GloVe work.

Graph Embedding Word Embeddings

Learning to Sketch with Shortcut Cycle Consistency

no code implementations CVPR 2018 Jifei Song, Kaiyue Pang, Yi-Zhe Song, Tao Xiang, Timothy Hospedales

In this paper, we present a novel approach for translating an object photo to a sketch, mimicking the human sketching process.

Multi-Task Learning Retrieval +1

Transductive Zero-Shot Action Recognition by Word-Vector Embedding

no code implementations13 Nov 2015 Xun Xu, Timothy Hospedales, Shaogang Gong

This is a more challenging problem than existing ZSL of still images and/or attributes, because the mapping between video spacetime features of actions and the semantic space is more complex and harder to learn for the purpose of generalising over any cross-category domain shift.

Action Recognition Temporal Action Localization +1

Free-hand Sketch Synthesis with Deformable Stroke Models

no code implementations9 Oct 2015 Yi Li, Yi-Zhe Song, Timothy Hospedales, Shaogang Gong

We present a generative model which can automatically summarize the stroke composition of free-hand sketches of a given category.

Zero-Shot Domain Adaptation via Kernel Regression on the Grassmannian

no code implementations28 Jul 2015 Yongxin Yang, Timothy Hospedales

Most visual recognition methods implicitly assume the data distribution remains unchanged from training to testing.

regression Unsupervised Domain Adaptation

Discovery of Shared Semantic Spaces for Multi-Scene Video Query and Summarization

no code implementations27 Jul 2015 Xun Xu, Timothy Hospedales, Shaogang Gong

The growing rate of public space CCTV installations has generated a need for automated methods for exploiting video surveillance data including scene understanding, query, behaviour annotation and summarization.

Scene Understanding Semantic Similarity +2

Making Better Use of Edges via Perceptual Grouping

no code implementations CVPR 2015 Yonggang Qi, Yi-Zhe Song, Tao Xiang, Honggang Zhang, Timothy Hospedales, Yi Li, Jun Guo

We propose a perceptual grouping framework that organizes image edges into meaningful structures and demonstrate its usefulness on various computer vision tasks.

Learning-To-Rank Retrieval +1

When Face Recognition Meets with Deep Learning: an Evaluation of Convolutional Neural Networks for Face Recognition

no code implementations9 Apr 2015 Guosheng Hu, Yongxin Yang, Dong Yi, Josef Kittler, William Christmas, Stan Z. Li, Timothy Hospedales

In this work, we conduct an extensive evaluation of CNN-based face recognition systems (CNN-FRS) on a common ground to make our work easily reproducible.

Face Recognition Metric Learning

Semantic Embedding Space for Zero-Shot Action Recognition

no code implementations5 Feb 2015 Xun Xu, Timothy Hospedales, Shaogang Gong

In this framework a mapping is constructed between visual features and a human interpretable semantic description of each category, allowing categories to be recognised in the absence of any training data.

Action Recognition Data Augmentation +2

Sketch-a-Net that Beats Humans

2 code implementations30 Jan 2015 Qian Yu, Yongxin Yang, Yi-Zhe Song, Tao Xiang, Timothy Hospedales

We propose a multi-scale multi-channel deep neural network framework that, for the first time, yields sketch recognition performance surpassing that of humans.

Sketch Recognition

Cannot find the paper you are looking for? You can Submit a new open access paper.