Research feeds | TransferLab — appliedAI Institute

Research feeds

Staying abreast in the fast-paced world of machine learning research is hard. We contribute to the community by writing paper pills: summaries of publications, talks or new software, with context and an explanation of the main ideas and key results.

Paper pills are written for practitioners or researchers who want a peek into recent developments but are short on time. We publish monthly digests as well as occasional highlights of major conferences.

DataInf: Efficiently Estimating Data Influence in LoRA-tuned LLMs and Diffusion Models

Accurately computing influence functions involves solving inverse Hessian problems, a challenging task as the parameter count increases, …

Data-OOB: Out-of-bag Estimate as a Simple and Efficient Data Value

The out-of-bag (OOB) error estimate is a scalable approach to data valuation. Unlike marginal contribution methods, Data-OOB can leverage …

Exploiting past success in Off-Policy Actor-Critic

Extracting knowledge from previously gathered data is the very core of exploitation in reinforcement learning (RL). During training, a …

Studying Large Language Model Generalization with Influence Functions

Influence functions are a tool to quantify the impact of each training sample on a model&rsquo;s predictions, thereby assisting in the …

Neural Operators: Learning Maps Between Function Spaces

Neural operators are generalized neural networks that can map between infinite dimensional function spaces. A recent paper proposes a …

Modeling Dynamical Systems with Geometric Clifford Algebra Networks

The paper investigates the use of Clifford algebras in deep learning. The authors define a general parametric group action layer, which …

Random Sum-Product Networks: A Simple and Effective Approach to Probabilistic Deep Learning

Is structure learning even necessary? By generating overparametrised, random model structures for sum-product networks, a class of tractable …

Automatic Posterior Transformation for Likelihood-free Inference

A sequential neural posterior estimation method that modifies the posterior approximation using arbitrary, dynamically updated proposals. …

Jump-Start Reinforcement Learning

Reinforcement learning often presents a challenge when training an agent from scratch, as reward sparsity and exploration issues can impede …

Implicit Q Learning

A core challenge when applying dynamic programming based approaches to offline reinforcement learning is the bootstrapping error. This pill …

Deep Learning with Functional Inputs

Functional data is common in many applications but has not been included into deep-learning methods naturally. This paper presents an input …

Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training

The process of pre-training large language models can incur significant expenses. As these models continue to grow in size, there is an …

Learning-rate-free learning by D-Adaptation

Fine-tuning the learning rate (lr) in the training of neural networks is crucial for their performance, and often requires a costly search …

Reasoning Traces as Learning Signal

An important feature of large language models is their ability to provide detailed responses that resemble &ldquo;thinking step by …

Advantage-Induced Policy Alignment

Building on the classic results on reward weighted regression and its more recent adaptation to deep learning, a new algorithm called …

Direct Preference Optimization

With direct preference optimization (DPO), a language model can be aligned with human preferences without using reinforcement learning, …

Augmented Language Models: a survey

A survey of recent advances in augmenting (large) language models with new capabilities such as reasoning, tool use, and more. While the …

Joint Probability Trees

Joint probability trees (JPTs) are a novel formalism for the representation of full-joint distributions over sets of random variables in …

Density estimation using deep generative neural networks

Density estimation is among the fundamental problems in statistics. It is difficult to estimate the density of high-dimensional data due to …

Physics-informed DeepONets

Physics-informed DeepONets can learn the solution operator of arbitrary partial differential equations, even without any paired …

Influence functions are a tool to quantify the impact of each training sample on a model’s predictions, thereby assisting in the …

An important feature of large language models is their ability to provide detailed responses that resemble “thinking step by …