multilayer perceptron
PulseAugur coverage of multilayer perceptron — every cluster mentioning multilayer perceptron across labs, papers, and developer communities, ranked by signal.
- instance of Gotit.pub 95%
- instance of IArxiv 90%
- instance of IArxiv Recommender 90%
- used by Gotit.pub 70%
- instance of Influence Flower 70%
- used by Influence Flower 70%
- used by CatalyzeX Code Finder for Papers 70%
- used by magnetic resonance imaging 70%
- instance of CatalyzeX Code Finder for Papers 70%
- instance of grokking 70%
- used by AlphaEarth 70%
- competes with Diffusion Transformer 70%
19 day(s) with sentiment data
-
New CJRR-Net method enhances human pose estimation via viewpoint selection
Researchers have developed a new method called the Candidate-Joint Reprojection Ranking Network (CJRR-Net) to improve human pose estimation in multi-camera setups. This network selects the most informative viewpoint by …
-
FUTON: New Fourier Tensor Networks Enhance Implicit Neural Representations
Researchers have introduced FUTON (Fourier Tensor Network), a novel approach to implicit neural representations (INRs). FUTON models signals as generalized Fourier series, utilizing low-rank tensor decomposition to para…
-
Data diversity, not architecture, drives symbolic mechanisms in LLMs
A new research paper investigates the emergence of symbolic mechanisms in large language models, finding that data diversity, rather than architectural design, is the primary driver. Researchers trained neural networks …
-
Neural network weight evolution analyzed using topological dynamics
Researchers have explored the topological dynamics of neural network weights under different supervision conditions. By analyzing persistent homology of weight point clouds in a multilayer perceptron trained on Fashion-…
-
New HOPE method prunes MoE models by considering expert interactions
Researchers have developed HOPE (Higher-Order Pruning of Experts), a novel method for reducing the memory footprint of Mixture-of-Experts (MoE) language models. Unlike previous first-order methods that assess experts in…
-
New method automatically discovers symbolic local learning rules for deep learning
Researchers have developed a novel method for automatically discovering symbolic local learning rules that could potentially replace backpropagation in deep learning architectures. This technique involves creating a dic…
-
Tabular Foundation Models Show Promise as Training-Free World Models
A new paper explores the potential of pretrained tabular foundation models (TabFM) to act as training-free world models for reinforcement learning and robot planning. The research found that TabFM outperformed models li…
-
New AI model identifies optimal global solar power plant locations
Researchers have developed a new method for identifying optimal locations for solar power plants using geospatial embeddings. This approach analyzes existing solar sites to learn a suitability signal, outperforming trad…
-
Qwen3.5 model neurons show unexpected, structured ordering
Researchers have discovered an unusual ordering pattern within the neurons of the Qwen3.5-0.8B model, where more active neurons are consistently found at lower indices within each layer. This phenomenon, observed in bot…
-
EEG foundation model adapters show no consistent advantage over simpler controls
A study published on arXiv explored the effectiveness of different adapter designs for electroencephalography (EEG) foundation models. Researchers compared a "backbone-informed" adapter, which models relationships acros…
-
LLM Math Errors Linked to Late-Network Sign Flips
A new study investigates why large language models make sign errors in multi-step mathematics, finding that these errors are committed late in the network, typically within the final quarter of MLP layers. The research …
-
AI accurately forecasts material grain growth under complex thermal conditions
Researchers have developed a new framework for forecasting grain growth in polycrystalline materials under non-isothermal conditions. This approach utilizes a combination of convolutional neural networks to encode grain…
-
New PEARL method enables selective forgetting in personalized AI models
Researchers have developed PEARL (Prediction Error-Adaptive Regularized Learning), a novel method for updating personalized AI models when user preferences change over time. Unlike existing methods that forget uniformly…
-
New FLAT method improves sample-efficient traffic calibration for digital twins
Researchers have developed FLAT, a novel method for sample-efficient traffic calibration in digital twins. This approach smooths the parameter-error landscape, making the objective learnable from a limited number of sam…
-
New research quantifies stochastic rounding impact on SGD training
A new arXiv paper explores the impact of stochastic rounding on the training of deep learning models using Stochastic Gradient Descent (SGD). The research provides theoretical bounds on prediction error for linear netwo…
-
Deep learning architectures rigorously benchmarked for financial forecasting · 2 sources tracked
Two research papers explore the application and evaluation of deep learning models for financial forecasting. The first paper introduces FinForesight, a diagnostic benchmark designed to rigorously test deep research sys…
-
Quantum ML models audited for fairness in medical imaging
Researchers have conducted a fairness audit on quantum machine learning (QML) models used for medical imaging, finding that while QML can achieve competitive performance, it may still exhibit disparities across sensitiv…
-
New research quantifies limits of LLM decoding length prediction
A new arXiv paper introduces a method for predicting the decoding length of large language models using only prefill-phase hidden states. The research highlights that current benchmarks often conflate predictor quality …
-
HECKTOR 2026 challenge uses AI for head and neck cancer prediction
The HECKTOR 2026 challenge focuses on machine learning applications for head and neck cancer, involving tumor delineation, TN staging, and recurrence-free survival prediction. One team, nucli, utilized a self-supervised…
-
New neural ODE method discovers Lie symmetries and first integrals from data
Researchers have developed LatentLieODE, a novel neural ordinary differential equation (ODE) method capable of identifying linear Lie symmetries and first integrals from observed time-series data, such as videos. This m…