Search Results for author:
Found 27 papers, 16 papers with code
Trajectory Regularization Enhances Self-Supervised Geometric Representation
To address this gap, we introduce a new pose-estimation benchmark for assessing SSL geometric representations, which demands training without semantic or pose labels and achieving proficiency in both semantic and geometric downstream tasks.
Gen4Gen: Generative Data Pipeline for Generative Multi-Concept Composition
First, current personalization techniques fail to reliably extend to multiple concepts -- we hypothesize this to be due to the mismatch between complex scenes and simple text descriptions in the pre-training dataset (e. g., LAION).
URLOST: Unsupervised Representation Learning without Stationarity or Topology
Unsupervised representation learning has seen tremendous progress but is constrained by its reliance on data modality-specific stationarity and topology, a limitation not found in biological intelligence systems.
Unsupervised Feature Learning with Emergent Data-Driven Prototypicality
With our feature mapper simply trained to spread out training instances in hyperbolic space, we observe that images move closer to the origin with congealing, validating our idea of unsupervised prototypicality discovery.
Variance-Covariance Regularization Improves Representation Learning
In summary, VCReg offers a universally applicable regularization framework that significantly advances transfer learning and highlights the connection between gradient starvation, neural collapse, and feature transferability.
EMP-SSL: Towards Self-Supervised Learning in One Training Epoch
Recently, self-supervised learning (SSL) has achieved tremendous success in learning image representation.
Unsupervised Learning of Structured Representations via Closed-Loop Transcription
This paper proposes an unsupervised method for learning a unified representation that serves both discriminative and generative purposes.
Simple Emergent Action Representations from Multi-Task Policy Training
The low-level sensory and motor signals in deep reinforcement learning, which exist in high-dimensional spaces such as image observations or motor torques, are inherently challenging to understand or utilize directly for downstream tasks.
Minimalistic Unsupervised Learning with the Sparse Manifold Transform
Though there remains a small performance gap between our simple constructive model and SOTA methods, the evidence points to this as a promising direction for achieving a principled and white-box approach to unsupervised learning.
Ranked #1 on
Unsupervised MNIST
on MNIST
Self-Supervised Learning
Sparse Representation-based Classification
+3
Joint Embedding Self-Supervised Learning in the Kernel Regime
The fundamental goal of self-supervised learning (SSL) is to produce useful representations of data without access to any labels for classifying the data.
Bag of Image Patch Embedding Behind the Success of Self-Supervised Learning
Even with 32x32 patch representation, BagSSL achieves 62% top-1 linear probing accuracy on ImageNet.
On the duality between contrastive and non-contrastive self-supervised learning
Recent approaches in self-supervised learning of image representations can be categorized into different families of methods and, in particular, can be divided into contrastive and non-contrastive approaches.
Neural Manifold Clustering and Embedding
We argue that achieving manifold clustering with neural networks requires two essential ingredients: a domain-specific constraint that ensures the identification of the manifolds, and a learning algorithm for embedding each manifold to a linear subspace in the feature space.
Decoupled Contrastive Learning
Further, DCL can be combined with the SOTA contrastive learning method, NNCLR, to achieve 72. 3% ImageNet-1K top-1 accuracy with 512 batch size in 400 epochs, which represents a new SOTA in contrastive learning.
Clipped Hyperbolic Classifiers Are Super-Hyperbolic Classifiers
Hyperbolic space can naturally embed hierarchies, unlike Euclidean space.
Compact and Optimal Deep Learning with Recurrent Parameter Generators
We propose a drastically different approach to compact and optimal deep learning: We decouple the Degrees of freedom (DoF) and the actual number of parameters of a model, optimize a small DoF with predefined random linear constraints for a large model of arbitrary architecture, in one-stage end-to-end learning.
Ranked #97 on
Image Classification
on ObjectNet
(using extra training data)
Transformer visualization via dictionary learning: contextualized embedding as a linear superposition of transformer factors
Transformer networks have revolutionized NLP representation learning since they were introduced.
Disentangling images with Lie group transformations and sparse coding
Discrete spatial patterns and their continuous transformations are two important regularities contained in natural signals.
A Neural Network MCMC sampler that maximizes Proposal Entropy
However, in the continuous case, unfavorable geometry of the target distribution can greatly limit the efficiency of MCMC methods.
RG-Flow: A hierarchical and explainable flow model based on renormalization group and sparse prior
In this work, we incorporate the key ideas of renormalization group (RG) and sparse prior distribution to design a hierarchical flow-based generative model, RG-Flow, which can separate information at different scales of images and extract disentangled representations at each scale.
3D Shape Reconstruction from Free-Hand Sketches
Additionally, we propose a sketch standardization module to handle different sketch distortions and styles.
Orthogonal Convolutional Neural Networks
We develop an efficient approach to impose filter orthogonality on a convolutional layer based on the doubly block-Toeplitz matrix representation of the convolutional kernel instead of using the common kernel orthogonality approach, which we show is only necessary but not sufficient for ensuring orthogonal convolutions.
Learning Energy-Based Models in High-Dimensional Spaces with Multi-scale Denoising Score Matching
Recently, \citet{song2019generative} have shown that a generative model trained by denoising score matching accomplishes excellent sample synthesis, when trained with data samples corrupted with multiple levels of noise.
Word Embedding Visualization Via Dictionary Learning
Co-occurrence statistics based word embedding techniques have proved to be very useful in extracting the semantic and syntactic representation of words as low dimensional continuous vectors.
Annealed Denoising score matching: learning Energy based model in high-dimensional spaces
Energy based models outputs unmormalized log-probability values given datasamples.
Superposition of many models into one
We present a method for storing multiple models within a single set of parameters.
The Sparse Manifold Transform
We present a signal representation framework called the sparse manifold transform that combines key ideas from sparse coding, manifold learning, and slow feature analysis.
