Mark Coates
YOU?
Author Swipe
PriviRec: Confidential and Decentralized Graph Filtering for Recommender Systems Open
International audience
View article: It Takes Two: Your GRPO Is Secretly DPO
It Takes Two: Your GRPO Is Secretly DPO Open
Group Relative Policy Optimization (GRPO) is a prominent reinforcement learning algorithm for post-training Large Language Models (LLMs). It is commonly believed that GRPO necessitates a large group size to ensure stable training via preci…
View article: Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling
Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling Open
The pursuit of general-purpose artificial intelligence depends on large language models (LLMs) that can handle both structured reasoning and open-ended generation. We present Omni-Thinker, a unified reinforcement learning (RL) framework th…
View article: Communication Efficient, Differentially Private Distributed Optimization using Correlation-Aware Sketching
Communication Efficient, Differentially Private Distributed Optimization using Correlation-Aware Sketching Open
Federated learning with differential privacy suffers from two major costs: each client must transmit $d$-dimensional gradients every round, and the magnitude of DP noise grows with $d$. Yet empirical studies show that gradient updates exhi…
SKOLR: Structured Koopman Operator Linear RNN for Time-Series Forecasting Open
Koopman operator theory provides a framework for nonlinear dynamical system analysis and time-series forecasting by mapping dynamics to a space of real-valued measurement functions, enabling a linear operator representation. Despite the ad…
View article: One Demo Is All It Takes: Planning Domain Derivation with LLMs from A Single Demonstration
One Demo Is All It Takes: Planning Domain Derivation with LLMs from A Single Demonstration Open
Pre-trained large language models (LLMs) show promise for robotic task planning but often struggle to guarantee correctness in long-horizon problems. Task and motion planning (TAMP) addresses this by grounding symbolic plans in low-level e…
Half Search Space is All You Need Open
Neural Architecture Search (NAS) is a powerful tool for automating architecture design. One-Shot NAS techniques, such as DARTS, have gained substantial popularity due to their combination of search efficiency with simplicity of implementat…
View article: Plain Transformers Can be Powerful Graph Learners
Plain Transformers Can be Powerful Graph Learners Open
Transformers have attained outstanding performance across various modalities, owing to their simple but powerful scaled-dot-product (SDP) attention mechanisms. Researchers have attempted to migrate Transformers to graph learning, but most …
Variation Matters: from Mitigating to Embracing Zero-Shot NAS Ranking Function Variation Open
Neural Architecture Search (NAS) is a powerful automatic alternative to manual design of a neural network. In the zero-shot version, a fast ranking function is used to compare architectures without training them. The outputs of the ranking…
View article: InnerThoughts: Disentangling Representations and Predictions in Large Language Models
InnerThoughts: Disentangling Representations and Predictions in Large Language Models Open
Large language models (LLMs) contain substantial factual knowledge which is commonly elicited by multiple-choice question-answering prompts. Internally, such models process the prompt through multiple transformer layers, building varying r…
View article: Secure Federated Graph-Filtering for Recommender Systems
Secure Federated Graph-Filtering for Recommender Systems Open
Recommender systems often rely on graph-based filters, such as normalized item-item adjacency matrices and low-pass filters. While effective, the centralized computation of these components raises concerns about privacy, security, and the …
View article: Path-of-Thoughts: Extracting and Following Paths for Robust Relational Reasoning with Large Language Models
Path-of-Thoughts: Extracting and Following Paths for Robust Relational Reasoning with Large Language Models Open
Large language models (LLMs) possess vast semantic knowledge but often struggle with complex reasoning tasks, particularly in relational reasoning problems such as kinship or spatial reasoning. In this paper, we present Path-of-Thoughts (P…
View article: Refining Answer Distributions for Improved Large Language Model Reasoning
Refining Answer Distributions for Improved Large Language Model Reasoning Open
Large Language Models (LLMs) have exhibited an impressive capability to perform reasoning tasks, especially if they are encouraged to generate a sequence of intermediate steps. Reasoning performance can be improved by suitably combining mu…
View article: Differentially private and decentralized randomized power method
Differentially private and decentralized randomized power method Open
The randomized power method has gained significant interest due to its simplicity and efficient handling of large-scale spectral analysis and recommendation tasks. However, its application to large datasets containing personal information …
Sparse Decomposition of Graph Neural Networks Open
Graph Neural Networks (GNN) exhibit superior performance in graph representation learning, but their inference cost can be high, due to an aggregation operation that can require a memory fetch for a very large number of nodes. This inferen…
Enhancing Click-through Rate Prediction in Recommendation Domain with Search Query Representation Open
Many platforms, such as e-commerce websites, offer both search and\nrecommendation services simultaneously to better meet users' diverse needs.\nRecommendation services suggest items based on user preferences, while search\nservices allow …
HardCore Generation: Generating Hard UNSAT Problems for Data Augmentation Open
Efficiently determining the satisfiability of a boolean equation -- known as the SAT problem for brevity -- is crucial in various industrial problems. Recently, the advent of deep learning methods has introduced significant potential for e…
View article: Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Data
Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Data Open
Despite recent advances in training and prompting strategies for Large Language Models (LLMs), these models continue to face challenges with complex logical reasoning tasks that involve long reasoning chains. In this work, we explore the p…
Predicting Probabilities of Error to Combine Quantization and Early Exiting: QuEE Open
Machine learning models can solve complex tasks but often require significant computational resources during inference. This has led to the development of various post-training computation reduction methods that tackle this issue in differ…
Graph Knowledge Distillation to Mixture of Experts Open
In terms of accuracy, Graph Neural Networks (GNNs) are the best architectural choice for the node classification task. Their drawback in real-world deployment is the latency that emerges from the neighbourhood processing operation. One sol…
MODL: Multilearner Online Deep Learning Open
Online deep learning tackles the challenge of learning from data streams by balancing two competing goals: fast learning and deep learning. However, existing research primarily emphasizes deep learning solutions, which are more adept at ha…
View article: GraSS: Combining Graph Neural Networks with Expert Knowledge for SAT Solver Selection
GraSS: Combining Graph Neural Networks with Expert Knowledge for SAT Solver Selection Open
Boolean satisfiability (SAT) problems are routinely solved by SAT solvers in real-life applications, yet solving time can vary drastically between solvers for the same instance. This has motivated research into machine learning models that…
View article: CKGConv: General Graph Convolution with Continuous Kernels
CKGConv: General Graph Convolution with Continuous Kernels Open
The existing definitions of graph convolution, either from spatial or spectral perspectives, are inflexible and not unified. Defining a general convolution operator in the graph domain is challenging due to the lack of canonical coordinate…
View article: Personalized Negative Reservoir for Incremental Learning in Recommender Systems
Personalized Negative Reservoir for Incremental Learning in Recommender Systems Open
Recommender systems have become an integral part of online platforms. Every day the volume of training data is expanding and the number of user interactions is constantly increasing. The exploration of larger and more expressive models has…
Population Monte Carlo with Normalizing Flow Open
Adaptive importance sampling (AIS) methods provide a useful alternative to Markov Chain Monte Carlo (MCMC) algorithms for performing inference of intractable distributions. Population Monte Carlo (PMC) algorithms constitute a family of AIS…
View article: Multi-resolution Time-Series Transformer for Long-term Forecasting
Multi-resolution Time-Series Transformer for Long-term Forecasting Open
The performance of transformers for time-series forecasting has improved significantly. Recent architectures learn complex temporal patterns by segmenting a time-series into patches and using the patches as tokens. The patch size controls …
Interacting Diffusion Processes for Event Sequence Forecasting Open
Neural Temporal Point Processes (TPPs) have emerged as the primary framework for predicting sequences of events that occur at irregular time intervals, but their sequential nature can hamper performance for long-horizon forecasts. To addre…
Jointly-Learned Exit and Inference for a Dynamic Neural Network : JEI-DNN Open
Large pretrained models, coupled with fine-tuning, are slowly becoming established as the dominant architecture in machine learning. Even though these models offer impressive performance, their practical application is often limited by the…
Substituting Data Annotation with Balanced Updates and Collective Loss in Multi-label Text Classification Open
Multi-label text classification (MLTC) is the task of assigning multiple labels to a given text, and has a wide range of application domains. Most existing approaches require an enormous amount of annotated data to learn a classifier and/o…
Neighbor Auto-Grouping Graph Neural Networks for Handover Parameter Configuration in Cellular Network Open
The mobile communication enabled by cellular networks is the one of the main foundations of our modern society. Optimizing the performance of cellular networks and providing massive connectivity with improved coverage and user experience h…