Search Results for author:
Found 101 papers, 65 papers with code
Towards a Robust Soft Baby Robot With Rich Interaction Ability for Advanced Machine Learning Algorithms
We believe one of the reasons for this is the disconnect between traditional robotic design and the properties needed for open-ended, creativity-based AI systems.
Cross-Attention Makes Inference Cumbersome in Text-to-Image Diffusion Models
This study explores the role of cross-attention during inference in text-conditional diffusion models.
Learning Useful Representations of Recurrent Neural Network Weight Matrices
The program of an RNN is its weight matrix.
Language Agents as Optimizable Graphs
Various human-designed prompt engineering techniques have been proposed to improve problem solvers based on Large Language Models (LLMs), yielding many disparate code bases.
SwitchHead: Accelerating Transformers with Mixture-of-Experts Attention
The costly self-attention layers in modern Transformers require memory and compute quadratic in sequence length.
Automating Continual Learning
General-purpose learning systems should improve themselves in open-ended fashion in ever-changing environments.
Efficient Rotation Invariance in Deep Neural Networks through Artificial Mental Rotation
Humans and animals recognize objects irrespective of the beholder's point of view, which may drastically change their appearances.
Unsupervised Musical Object Discovery from Audio
Our novel MusicSlots method adapts SlotAttention to the audio domain, to achieve unsupervised music decomposition.
Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions
Recent studies of the computational power of recurrent neural networks (RNNs) reveal a hierarchy of RNN architectures, given real-time and finite-precision assumptions.
Approximating Two-Layer Feedforward Networks for Efficient Transformers
Unlike prior work that compares MoEs with dense baselines under the compute-equal condition, our evaluation condition is parameter-equal, which is crucial to properly evaluate LMs.
The Languini Kitchen: Enabling Language Modelling Research at Different Scales of Compute
We introduce an experimental protocol that enables model comparisons based on equivalent compute, measured in accelerator hours.
Learning to Identify Critical States for Reinforcement Learning from Videos
Recent work on deep reinforcement learning (DRL) has pointed out that algorithmic information about good policies can be extracted from offline data which lack explicit information about executed actions.
MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs).
Ranked #7 on
Code Generation
on HumanEval
Exploring the Promise and Limits of Real-Time Recurrent Learning
To scale to such challenging tasks, we focus on certain well-known neural architectures with element-wise recurrence, allowing for tractable RTRL without approximation.
Mindstorms in Natural Language-Based Societies of Mind
What should be the social structure of an NLSOM?
Contrastive Training of Complex-Valued Autoencoders for Object Discovery
Current state-of-the-art object-centric models use slots and attention-based routing for binding.
Large Language Model Programs
In recent years, large pre-trained language models (LLMs) have demonstrated the ability to follow instructions and perform novel tasks from a few examples.
Accelerating Neural Self-Improvement via Bootstrapping
Few-shot learning with sequence-processing neural networks (NNs) has recently attracted a new wave of attention in the context of large language models.
Topological Neural Discrete Representation Learning à la Kohonen
Unsupervised learning of discrete representations from continuous ones in neural networks (NNs) is the cornerstone of several applications today.
Guiding Online Reinforcement Learning with Action-Free Offline Pretraining
In this paper, we investigate the potential of using action-free offline datasets to improve online reinforcement learning, name this problem Reinforcement Learning with Action-Free Offline Pretraining (AFP-RL).
Eliminating Meta Optimization Through Self-Referential Meta Learning
We discuss the relationship of such systems to in-context and memory-based meta learning and show that self-referential neural networks require functionality to be reused in the form of parameter sharing.
Learning One Abstract Bit at a Time Through Self-Invented Experiments Encoded as Neural Networks
There are two important things in science: (A) Finding answers to given questions, and (B) Coming up with good questions.
On Narrative Information and the Distillation of Stories
We then demonstrate how evolutionary algorithms can leverage this to extract a set of narrative templates and how these templates -- in tandem with a novel curve-fitting algorithm we introduce -- can reorder music albums to automatically induce stories in them.
Exploring through Random Curiosity with General Value Functions
Furthermore, RC-GVF significantly outperforms previous methods in the absence of ground-truth episodic counts in the partially observable MiniGrid environments.
Learning to Control Rapidly Changing Synaptic Connections: An Alternative Type of Memory in Sequence Processing Artificial Neural Networks
Short-term memory in standard, general-purpose, sequence-processing recurrent neural networks (RNNs) is stored as activations of nodes or "neurons."
The Benefits of Model-Based Generalization in Reinforcement Learning
First, we provide a simple theorem motivating how learning a model as an intermediate step can narrow down the set of possible value functions more than learning a value function directly from data using the Bellman equation.
Model-based Reinforcement Learning
reinforcement-learning
+1
CTL++: Evaluating Generalization on Never-Seen Compositional Patterns of Known Functions, and Compatibility of Neural Representations
While the original CTL is used to test length generalization or productivity, CTL++ is designed to test systematicity of NNs, that is, their capability to generalize to unseen compositions of known functions.
Images as Weight Matrices: Sequential Image Generation Through Synaptic Learning Rules
Work on fast weight programmers has demonstrated the effectiveness of key/value outer product-based learning rules for sequentially generating a weight matrix (WM) of a neural net (NN) by another NN or itself.
Learning to Generalize with Object-centric Agents in the Open World Survival Game Crafter
We show that current agents struggle to generalize, and introduce novel object-centric agents that improve over strong baselines.
Goal-Conditioned Generators of Deep Policies
A form of weight-sharing HyperNetworks and policy embeddings scales our method to generate deep NNs.
General Policy Evaluation and Improvement by Learning to Identify Few But Crucial States
In continuous control problems with infinitely many states, our value function minimizes its prediction error by simultaneously learning a small set of `probing states' and a mapping from actions produced in probing states to the policy's return.
Neural Differential Equations for Learning to Program Neural Nets Through Continuous Learning Rules
Neural ordinary differential equations (ODEs) have attracted much attention as continuous-time counterparts of deep residual neural networks (NNs), and numerous extensions for recurrent NNs have been proposed.
Upside-Down Reinforcement Learning Can Diverge in Stochastic Environments With Episodic Resets
Upside-Down Reinforcement Learning (UDRL) is an approach for solving RL problems that does not require value functions and uses only supervised learning, where the targets for given inputs in a dataset do not change over time.
Unsupervised Learning of Temporal Abstractions with Slot-based Transformers
The discovery of reusable sub-routines simplifies decision-making and planning in complex reinforcement learning problems.
All You Need Is Supervised Learning: From Imitation Learning to Meta-RL With Upside Down RL
Upside down reinforcement learning (UDRL) flips the conventional use of the return in the objective function in RL upside down, by taking returns as input and predicting actions.
Learning Relative Return Policies With Upside-Down Reinforcement Learning
Lately, there has been a resurgence of interest in using supervised learning to solve reinforcement learning problems.
The Dual Form of Neural Networks Revisited: Connecting Test Time Predictions to Training Patterns via Spotlights of Attention
Linear layers in neural networks (NNs) trained by gradient descent can be expressed as a key-value memory system which stores all training datapoints and the initial weights, and produces outputs using unnormalised dot attention over the entire training experience.
A Modern Self-Referential Weight Matrix That Learns to Modify Itself
The weight matrix (WM) of a neural network (NN) is its program.
Improving Baselines in the Wild
We share our experience with the recently released WILDS benchmark, a collection of ten datasets dedicated to developing models and training strategies which are robust to domain shifts.
Training and Generating Neural Networks in Compressed Weight Space
The inputs and/or outputs of some neural nets are weight matrices of other neural nets.
Automatic Embedding of Stories Into Collections of Independent Media
We look at how machine learning techniques that derive properties of items in a collection of independent media can be used to automatically embed stories into such collections.
The Neural Data Router: Adaptive Control Flow in Transformers Improves Systematic Generalization
Despite progress across a broad range of applications, Transformers have limited success in systematic generalization.
Learning Adaptive Control Flow in Transformers for Improved Systematic Generalization
Despite successes across a broad range of applications, Transformers have limited capability in systematic generalization.
Augmenting Classic Algorithms with Neural Components for Strong Generalisation on Ambiguous and High-Dimensional Data
We augment classic algorithms with learned components to adapt them to domains currently dominated by deep learning models.
Adaptive Control Flow in Transformers Improves Systematic Generalization
Despite successes across a broad range of applications, Transformers have limited capability in systematic generalization.
The Devil is in the Detail: Simple Tricks Improve Systematic Generalization of Transformers
Our models improve accuracy from 50% to 85% on the PCFG productivity split, and from 35% to 81% on COGS.
Reward-Weighted Regression Converges to a Global Optimum
Reward-Weighted Regression (RWR) belongs to a family of widely known iterative Reinforcement Learning algorithms based on the Expectation-Maximization framework.
Bayesian brains and the Rényi divergence
Under the Bayesian brain hypothesis, behavioural variations can be attributed to different priors over generative model parameters.
Going Beyond Linear Transformers with Recurrent Fast Weight Programmers
Transformers with linearised attention (''linear Transformers'') have demonstrated the practical scalability and effectiveness of outer product-based Fast Weight Programmers (FWPs) from the '90s.
Spatial Dependency Networks: Neural Layers for Improved Generative Image Modeling
We show that augmenting the decoder of a hierarchical VAE by spatial dependency layers considerably improves density estimation over baseline convolutional architectures and the state-of-the-art among the models within the same class.
Is it enough to optimize CNN architectures on ImageNet?
In this work we challenge the notion that CNN architecture design solely based on ImageNet leads to generally effective convolutional neural network (CNN) architectures that perform well on a diverse set of datasets and application domains.
Linear Transformers Are Secretly Fast Weight Programmers
We show the formal equivalence of linearised self-attention mechanisms and fast weight controllers from the early '90s, where a ``slow" neural net learns by gradient descent to program the ``fast weights" of another net through sequences of elementary programming instructions which are additive outer products of self-invented activation patterns (today called keys and values).
Meta Learning Backpropagation And Improving It
Many concepts have been proposed for meta learning with neural networks (NNs), e. g., NNs that learn to reprogram fast weights, Hebbian plasticity, learned learning rules, and meta recurrent NNs.
On the Binding Problem in Artificial Neural Networks
Contemporary neural networks still fall short of human-level generalization, which extends far beyond our direct experiences.
Unsupervised Object Keypoint Learning using Local Spatial Predictability
We propose PermaKey, a novel approach to representation learning based on object keypoints.
Learning Associative Inference Using Fast Weight Memory
Humans can quickly associate stimuli to solve problems in novel contexts.
Ranked #1 on
Question Answering
on catbAbI LM-mode
Hierarchical Relational Inference
Common-sense physical reasoning in the real world requires learning about the interactions of objects and their dynamics.
Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks
Neural networks (NNs) whose subnetworks implement reusable functions are expected to offer numerous advantages, including compositionality through efficient recombination of functional building blocks, interpretability, preventing catastrophic interference, etc.
Recurrent Neural-Linear Posterior Sampling for Nonstationary Contextual Bandits
An agent in a nonstationary contextual bandit problem should balance between exploration and the exploitation of (periodic or structured) patterns present in its previous experiences.
Parameter-Based Value Functions
We introduce a class of value functions called Parameter-Based Value Functions (PBVFs) whose inputs include the policy parameters.
Meta-Learning for Recalibration of EMG-Based Upper Limb Prostheses
An EMG-based upper limb prosthesis relies on a statistical pattern recognition system to map the EMG signal of residual forearm muscles into the appropriate hand movements.
Training Agents using Upside-Down Reinforcement Learning
Many of its general principles are outlined in a companion report; the goal of this paper is to develop a practical learning algorithm and show that this conceptually simple perspective on agent training can produce a range of rewarding behaviors for multiple episodic environments.
Enhancing the Transformer with Explicit Relational Encoding for Math Problem Solving
We incorporate Tensor-Product Representations within the Transformer in order to better support the explicit representation of relation structure.
Ranked #1 on
Question Answering
on Mathematics Dataset
R-SQAIR: Relational Sequential Attend, Infer, Repeat
Traditional sequential multi-object attention models rely on a recurrent mechanism to infer object relations.
Improving Generalization in Meta Reinforcement Learning using Learned Objectives
Biological evolution has distilled the experiences of many learners into the general learning algorithms of humans.
Recurrent Neural Processes
We extend Neural Processes (NPs) to sequential data through Recurrent NPs or RNPs, a family of conditional state space models.
A Perspective on Objects and Systematic Generalization in Model-Based RL
In order to meet the diverse challenges in solving many real-world problems, an intelligent agent has to be able to dynamically construct a model of its environment.
Are Disentangled Representations Helpful for Abstract Visual Reasoning?
A disentangled representation encodes information about the salient factors of variation in the data independently.
Improving Differentiable Neural Computers Through Memory Masking, De-allocation, and Link Distribution Sharpness Control
The Differentiable Neural Computer (DNC) can learn algorithmic and question answering tasks.
Learning to Reason with Third Order Tensor Products
We combine Recurrent Neural Networks with Tensor Product Representations to learn combinatorial representations of sequential data.
Learning to Reason with Third-Order Tensor Products
We combine Recurrent Neural Networks with Tensor Product Representations to learn combinatorial representations of sequential data.
Investigating Object Compositionality in Generative Adversarial Networks
We present a minimal modification of a standard generator to incorporate this inductive bias and find that it reliably learns to generate images as compositions of objects.
Recurrent World Models Facilitate Policy Evolution
A generative recurrent neural network is quickly trained in an unsupervised manner to model popular reinforcement learning environments through compressed spatio-temporal representations.
World Models
We explore building generative neural network models of popular reinforcement learning environments.
DeepScores -- A Dataset for Segmentation, Detection and Classification of Tiny Objects
We present the DeepScores dataset with the goal of advancing the state-of-the-art in small objects recognition, and by placing the question of object recognition in the context of scene understanding.
Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions
Common-sense physical reasoning is an essential ingredient for any intelligent agent operating in the real-world.
GATED FAST WEIGHTS FOR ASSOCIATIVE RETRIEVAL
We improve previous end-to-end differentiable neural networks (NNs) with fast weight memories.
Neural Expectation Maximization
Many real world tasks such as reasoning and physical interaction require identification and manipulation of conceptual entities.
Highway and Residual Networks learn Unrolled Iterative Estimation
We demonstrate that this viewpoint directly leads to the construction of Highway and Residual networks.
Recurrent Highway Networks
We introduce a novel theoretical analysis of recurrent networks based on Gersgorin's circle theorem that illuminates several modeling and optimization issues and improves our understanding of the LSTM cell.
Ranked #16 on
Language Modelling
on Hutter Prize
Tagger: Deep Unsupervised Perceptual Grouping
We present a framework for efficient perceptual inference that explicitly reasons about the segmentation of its inputs and features.
Lipreading with Long Short-Term Memory
Lipreading, i. e. speech recognition from visual-only recordings of a speaker's face, can be achieved with a processing pipeline based solely on neural networks, yielding significantly better accuracy than conventional methods.
Binding via Reconstruction Clustering
Disentangled distributed representations of data are desirable for machine learning, since they are more expressive and can generalize from fewer examples.
Training Very Deep Networks
Theoretical and empirical evidence indicates that the depth of neural networks is crucial for their success.
Ranked #34 on
Image Classification
on MNIST
Highway Networks
There is plenty of theoretical and empirical evidence that depth of neural networks is a crucial ingredient for their success.
LSTM: A Search Space Odyssey
Several variants of the Long Short-Term Memory (LSTM) architecture for recurrent neural networks have been proposed since its inception in 1995.
Assessment of algorithms for mitosis detection in breast cancer histopathology images
The proliferative activity of breast tumors, which is routinely estimated by counting of mitotic figures in hematoxylin and eosin stained histology sections, is considered to be one of the most important prognostic markers.
Understanding Locally Competitive Networks
Recently proposed neural network activation functions such as rectified linear, maxout, and local winner-take-all have allowed for faster and more effective training of deep neural architectures on large and complex datasets.
A Clockwork RNN
Sequence prediction and classification are ubiquitous and challenging problems in machine learning that can require identifying complex dependencies between temporally distant inputs.
My First Deep Learning System of 1991 + Deep Learning Timeline 1962-2013
Deep Learning has attracted significant attention in recent years.
Compete to Compute
Local competition among neighboring neurons is common in biological neural networks (NNs).
Multi-Column Deep Neural Networks for Offline Handwritten Chinese Character Classification
Our Multi-Column Deep Neural Networks achieve best known recognition rates on Chinese characters from the ICDAR 2011 and 2013 offline handwriting competitions, approaching human performance.
Fast Image Scanning with Deep Max-Pooling Convolutional Neural Networks
Deep Neural Networks now excel at image classification, detection and segmentation.
Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images
The input layer maps each window pixel to a neuron.
Measuring Intelligence through Games
Artificial general intelligence (AGI) refers to research aimed at tackling the full problem of artificial intelligence, that is, create truly intelligent agents.
Natural Evolution Strategies
This paper presents Natural Evolution Strategies (NES), a recent family of algorithms that constitute a more principled approach to black-box optimization than established evolutionary algorithms.
High-Performance Neural Networks for Visual Object Classification
We present a fast, fully parameterizable GPU implementation of Convolutional Neural Network variants.
Improving the Asymptotic Performance of Markov Chain Monte-Carlo by Inserting Vortices
We present a new way of converting a reversible finite Markov chain into a nonreversible one, with a theoretical guarantee that the asymptotic variance of the MCMC estimator based on the non-reversible chain is reduced.
Offline Handwriting Recognition with Multidimensional Recurrent Neural Networks
Offline handwriting recognition---the transcription of images of handwritten text---is an interesting task, in that it combines computer vision with sequence learning.
Unconstrained On-line Handwriting Recognition with Recurrent Neural Networks
On-line handwriting recognition is unusual among sequence labelling tasks in that the underlying generator of the observed data, i. e. the movement of the pen, is recorded directly.
Connectionist Temporal Classification: Labelling Unsegmented Sequence Data with Recurrent Neural Networks
Many real-world sequence learning tasks require the prediction of sequences of labels from noisy, unsegmented input data.
