Search Results for author:
Found 99 papers, 38 papers with code
VL-ICL Bench: The Devil in the Details of Benchmarking Multimodal In-Context Learning
Built on top of LLMs, vision large language models (VLLMs) have advanced significantly in areas such as recognition, reasoning, and grounding.
SketchINR: A First Look into Sketches as Implicit Neural Representations
(ii) SketchINR's auto-decoder provides a much higher-fidelity representation than other learned vector sketch representations, and is uniquely able to scale to complex vector sketches such as FS-COCO.
Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models
Our experiments demonstrate that integrating this dataset into standard vision-language fine-tuning or utilizing it for post-hoc fine-tuning effectively safety aligns VLLMs.
On the Transferability of Large-Scale Self-Supervision to Few-Shot Audio Classification
In recent years, self-supervised learning has excelled for its capacity to learn robust feature representations from unlabelled data.
DemoFusion: Democratising High-Resolution Image Generation With No $$$
High-resolution image generation with Generative Artificial Intelligence (GenAI) has immense potential but, due to the enormous capital investment required for training, it is increasingly centralised to a few large corporations, and hidden behind paywalls.
Sketch-based Video Object Segmentation: Benchmark and Analysis
Reference-based video object segmentation is an emerging topic which aims to segment the corresponding target object in each video frame referred by a given reference, such as a language expression or a photo mask.
Is Scaling Learned Optimizers Worth It? Evaluating The Value of VeLO's 4000 TPU Months
We analyze VeLO (versatile learned optimizer), the largest scale attempt to train a general purpose "foundational" optimizer to date.
FairTune: Optimizing Parameter Efficient Fine Tuning for Fairness in Medical Image Analysis
We demonstrate empirically that FairTune leads to improved fairness on a range of medical imaging datasets.
Fool Your (Vision and) Language Model With Embarrassingly Simple Permutations
Large language and vision-language models are rapidly being deployed in practice thanks to their impressive capabilities in instruction following, in-context learning, and so on.
BayesDLL: Bayesian Deep Learning Library
We release a new Bayesian neural network library for PyTorch for large-scale deep networks.
Feed-Forward Source-Free Domain Adaptation via Class Prototypes
Source-free domain adaptation has become popular because of its practical usefulness and no need to access source data.
Label Calibration for Semantic Segmentation Under Domain Shift
Performance of a pre-trained semantic segmentation model is likely to substantially decrease on data from a new domain.
Evaluating the Evaluators: Are Current Few-Shot Learning Benchmarks Fit for Purpose?
Numerous benchmarks for Few-Shot Learning have been proposed in the last decade.
Navigating Noise: A Study of How Noise Influences Generalisation and Calibration of Neural Networks
Enhancing the generalisation abilities of neural networks (NNs) through integrating noise such as MixUp or Dropout during training has emerged as a powerful and adaptable technique.
A Hierarchical Bayesian Model for Deep Few-Shot Meta Learning
We propose a novel hierarchical Bayesian model for learning with a large (possibly infinite) number of tasks/episodes, which suits well the few-shot meta learning problem.
Neural Fine-Tuning Search for Few-Shot Learning
In few-shot recognition, a classifier that has been trained on one set of classes is required to rapidly adapt and generalize to a disjoint, novel set of classes.
Parameter-Efficient Fine-Tuning for Medical Image Analysis: The Missed Opportunity
We present a comprehensive evaluation of Parameter-Efficient Fine-Tuning (PEFT) techniques for diverse medical image analysis tasks.
Meta Omnium: A Benchmark for General-Purpose Learning-to-Learn
Meta-learning and other approaches to few-shot learning are widely studied for image recognition, and are increasingly applied to other vision tasks such as pose estimation and dense prediction.
FedHB: Hierarchical Bayesian Federated Learning
We propose a novel hierarchical Bayesian approach to Federated Learning (FL), where our model reasonably describes the generative process of clients' local data via hierarchical Bayesian modeling: constituting random variables of local models for clients that are governed by a higher-level global variate.
Fairness in AI and Its Long-Term Implications on Society
Successful deployment of artificial intelligence (AI) in various settings has led to numerous positive outcomes for individuals and society.
ChiroDiff: Modelling chirographic data with Diffusion Models
Such strictly-ordered discrete factorization however falls short of capturing key properties of chirographic data -- it fails to build holistic understanding of the temporal concept due to one-way visibility (causality).
Self-Supervised Multimodal Learning: A Survey
In this survey, we provide a comprehensive review of the state-of-the-art in SSML, in which we elucidate three major challenges intrinsic to self-supervised learning with multimodal data: (1) learning representations from multimodal data without labels, (2) fusion of different modalities, and (3) learning with unaligned data.
Zero-Shot Everything Sketch-Based Image Retrieval, and in Explainable Style
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'').
Amortised Invariance Learning for Contrastive Self-Supervision
Contrastive self-supervised learning methods famously produce high quality transferable representations by learning invariances to different data augmentations.
Domain Generalisation via Domain Adaptation: An Adversarial Fourier Amplitude Approach
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.
Quality Diversity for Visual Pre-Training
Notably, the augmentations used in both supervised and self-supervised training lead to features with high invariance to spatial and appearance transformations.
On-the-Fly Category Discovery
Our code is available at https://github. com/PRIS-CV/On-the-fly-Category-Discovery.
Task-aware Adaptive Learning for Cross-domain Few-shot Learning
In this paper, we first observe the dependence of task-specific parameter configuration on the target task.
An Erudite Fine-Grained Visual Classification Model
Therefore, we first propose a feature disentanglement module and a feature re-fusion module to reduce negative transfer and boost positive transfer between different datasets.
Meta-Learned Kernel For Blind Super-Resolution Kernel Estimation
Recent image degradation estimation methods have enabled single-image super-resolution (SR) approaches to better upsample real-world images.
Federated Learning for Inference at Anytime and Anywhere
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.
MEDFAIR: Benchmarking Fairness for Medical Imaging
In this work, we introduce MEDFAIR, a framework to benchmark the fairness of machine learning models for medical imaging.
Attacking Adversarial Defences by Smoothing the Loss Landscape
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
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.
Feed-Forward Latent Domain Adaptation
Recognizing that device's data are likely to come from multiple latent domains that include a mixture of unlabelled domain-relevant and domain-irrelevant examples, we focus on the comparatively under-studied problem of latent domain adaptation.
Lessons learned from the NeurIPS 2021 MetaDL challenge: Backbone fine-tuning without episodic meta-learning dominates for few-shot learning image classification
Although deep neural networks are capable of achieving performance superior to humans on various tasks, they are notorious for requiring large amounts of data and computing resources, restricting their success to domains where such resources are available.
MetaAudio: A Few-Shot Audio Classification Benchmark
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.
Ranked #1 on
Few-Shot Audio Classification
on NSynth
Finding lost DG: Explaining domain generalization via model complexity
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.
Gaussian Process Meta Few-shot Classifier Learning via Linear Discriminant Laplace Approximation
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
We propose defensive tensorization, an adversarial defence technique that leverages a latent high-order factorization of the network.
Online Hyperparameter Meta-Learning with Hypergradient Distillation
Many gradient-based meta-learning methods assume a set of parameters that do not participate in inner-optimization, which can be considered as hyperparameters.
SketchODE: Learning neural sketch representation in continuous time
Learning meaningful representations for chirographic drawing data such as sketches, handwriting, and flowcharts is a gateway for understanding and emulating human creative expression.
Loss Function Learning for Domain Generalization by Implicit Gradient
We take a different approach, and explore the impact of the ERM loss function on out-of-domain generalisation.
Visual Representation Learning over Latent Domains
A fundamental shortcoming of deep neural networks is their specialization to a single task and domain.
Vision-based system identification and 3D keypoint discovery using dynamics constraints
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.
A Channel Coding Benchmark for Meta-Learning
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
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.
Meta-Calibration: Learning of Model Calibration Using Differentiable Expected Calibration Error
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
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
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.
Ranked #6 on
Few-Shot Image Classification
on Meta-Dataset
cross-domain few-shot learning
Few-Shot Image Classification
Reducing Implicit Bias in Latent Domain Learning
A fundamental shortcoming of deep neural networks is their specialization to a single task and domain.
Robust Domain Randomised Reinforcement Learning through Peer-to-Peer Distillation
In reinforcement learning, domain randomisation is an increasingly popular technique for learning more general policies that are robust to domain-shifts at deployment.
Margin-Based Transfer Bounds for Meta Learning with Deep Feature Embedding
By transferring knowledge learned from seen/previous tasks, meta learning aims to generalize well to unseen/future tasks.
Weight-Covariance Alignment for Adversarially Robust Neural Networks
Stochastic Neural Networks (SNNs) that inject noise into their hidden layers have recently been shown to achieve strong robustness against adversarial attacks.
Learning to Generate Novel Domains for Domain Generalization
This explicitly increases the diversity of available training domains and leads to a more generalizable model.
Ranked #65 on
Domain Generalization
on PACS
Learning the Prediction Distribution for Semi-Supervised Learning with Normalising Flows
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.
BézierSketch: A generative model for scalable vector sketches
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.
Flexible Dataset Distillation: Learn Labels Instead of Images
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.
A Probabilistic Model for Discriminative and Neuro-Symbolic Semi-Supervised Learning
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 $p(y|x)$.
NewtonianVAE: Proportional Control and Goal Identification from Pixels via Physical Latent Spaces
Learning low-dimensional latent state space dynamics models has been a powerful paradigm for enabling vision-based planning and learning for control.
Latent Domain Learning with Dynamic Residual Adapters
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.
ALBA : Reinforcement Learning for Video Object Segmentation
We treat this as a grouping problem by exploiting object proposals and making a joint inference about grouping over both space and time.
DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics
Robots are still limited to controlled conditions, that the robot designer knows with enough details to endow the robot with the appropriate models or behaviors.
SimpleMKKM: Simple Multiple Kernel K-means
We propose a simple yet effective multiple kernel clustering algorithm, termed simple multiple kernel k-means (SimpleMKKM).
Meta-Learning in Neural Networks: A Survey
We survey promising applications and successes of meta-learning such as few-shot learning and reinforcement learning.
Online Meta-Learning for Multi-Source and Semi-Supervised Domain Adaptation
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
+2
Sequential Learning for Domain Generalization
In DG this means encountering a sequence of domains and at each step training to maximise performance on the next domain.
Ranked #76 on
Domain Generalization
on PACS
Deep Domain-Adversarial Image Generation for Domain Generalisation
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.
Ranked #63 on
Domain Generalization
on PACS
Incremental Few-Shot Object Detection
To this end we propose OpeN-ended Centre nEt (ONCE), a detector designed for incrementally learning to detect novel class objects with few examples.
DADA: Differentiable Automatic Data Augmentation
In this paper, we propose Differentiable Automatic Data Augmentation (DADA) which dramatically reduces the cost.
Ranked #15 on
Data Augmentation
on ImageNet
Adversarial Generation of Informative Trajectories for Dynamics System Identification
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
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
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.
Simple and Effective Stochastic Neural Networks
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.
Zero-Shot Crowd Behavior Recognition
To that end, we relax the common assumption that each individual crowd video instance is only associated with a single crowd attribute.
Factorized Higher-Order CNNs with an Application to Spatio-Temporal Emotion Estimation
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.
Physics-as-Inverse-Graphics: Unsupervised Physical Parameter Estimation from Video
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.
Multi-relational Poincaré Graph Embeddings
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.
Ranked #38 on
Link Prediction
on WN18RR
Analogies Explained: Towards Understanding Word Embeddings
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.
Behavioural Repertoire via Generative Adversarial Policy Networks
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
Word translation, or bilingual dictionary induction, is an important capability that impacts many multilingual language processing tasks.
Deep Multi-Task Learning to Recognise Subtle Facial Expressions of Mental States
To advance subtle expression recognition, we contribute a Large-scale Subtle Emotions and Mental States in the Wild database (LSEMSW).
Meta-Learning Transferable Active Learning Policies by Deep Reinforcement Learning
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.
What the Vec? Towards Probabilistically Grounded Embeddings
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.
Learning to Sketch with Shortcut Cycle Consistency
In this paper, we present a novel approach for translating an object photo to a sketch, mimicking the human sketching process.
Deep Stock Representation Learning: From Candlestick Charts to Investment Decisions
We propose a novel investment decision strategy (IDS) based on deep learning.
Computational Finance
Learning to Learn: Meta-Critic Networks for Sample Efficient Learning
We propose a novel and flexible approach to meta-learning for learning-to-learn from only a few examples.
Deep Multi-task Representation Learning: A Tensor Factorisation Approach
Our approach applies to both homogeneous and heterogeneous MTL.
Frankenstein: Learning Deep Face Representations using Small Data
To train such networks, very large training sets are needed with millions of labeled images.
Transductive Zero-Shot Action Recognition by Word-Vector Embedding
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.
Free-hand Sketch Synthesis with Deformable Stroke Models
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
Most visual recognition methods implicitly assume the data distribution remains unchanged from training to testing.
Discovery of Shared Semantic Spaces for Multi-Scene Video Query and Summarization
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.
Making Better Use of Edges via Perceptual Grouping
We propose a perceptual grouping framework that organizes image edges into meaningful structures and demonstrate its usefulness on various computer vision tasks.
When Face Recognition Meets with Deep Learning: an Evaluation of Convolutional Neural Networks for Face Recognition
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.
Semantic Embedding Space for Zero-Shot Action Recognition
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.
Ranked #31 on
Zero-Shot Action Recognition
on UCF101
Sketch-a-Net that Beats Humans
We propose a multi-scale multi-channel deep neural network framework that, for the first time, yields sketch recognition performance surpassing that of humans.
A Survey on Heterogeneous Face Recognition: Sketch, Infra-red, 3D and Low-resolution
Heterogeneous face recognition (HFR) refers to matching face imagery across different domains.
