English
Related papers

Related papers: Softly Induced Functional Simplicity: Implications…

200 papers

Recent works have demonstrated that neural networks exhibit extreme simplicity bias(SB). That is, they learn only the simplest features to solve a task at hand, even in the presence of other, more robust but more complex features. Due to…

Machine Learning · Computer Science 2023-02-02 Depen Morwani , Jatin Batra , Prateek Jain , Praneeth Netrapalli

Although recent works have brought some insights into the performance improvement of techniques used in state-of-the-art deep-learning models, more work is needed to understand their generalization properties. We shed light on this matter…

Machine Learning · Computer Science 2020-07-31 Mahsa Forouzesh , Farnood Salehi , Patrick Thiran

By representing words with probability densities rather than point vectors, probabilistic word embeddings can capture rich and interpretable semantic information and uncertainty. The uncertainty information can be particularly meaningful in…

Computation and Language · Computer Science 2018-04-30 Ben Athiwaratkun , Andrew Gordon Wilson

Recent work in scientific machine learning has developed so-called physics-informed neural network (PINN) models. The typical approach is to incorporate physical domain knowledge as soft constraints on an empirical loss function and use…

Machine Learning · Computer Science 2021-11-12 Aditi S. Krishnapriyan , Amir Gholami , Shandian Zhe , Robert M. Kirby , Michael W. Mahoney

Physical symmetries provide a strong inductive bias for constructing functions to analyze data. In particular, this bias may improve robustness, data efficiency, and interpretability of machine learning models. However, building machine…

High Energy Physics - Phenomenology · Physics 2025-11-05 Pradyun Hebbar , Thandikire Madula , Vinicius Mikuni , Benjamin Nachman , Nadav Outmezguine , Inbar Savoray

Decision-making in complex systems often relies on machine learning models, yet highly accurate models such as XGBoost and neural networks can obscure the reasoning behind their predictions. In operations research applications,…

Machine Learning · Computer Science 2025-02-28 Gaurav Arwade , Sigurdur Olafsson

Reinforcement Learning (RL) can enable agents to learn complex tasks. However, it is difficult to interpret the knowledge and reuse it across tasks. Inductive biases can address such issues by explicitly providing generic yet useful…

Artificial Intelligence · Computer Science 2022-12-13 Thomas Schnürer , Malte Probst , Horst-Michael Gross

Issues of safety, explainability, and efficiency are of increasing concern in learning systems deployed with hard and soft constraints. Symbolic Constrained Learning and Knowledge Distillation techniques have shown promising results in this…

Artificial Intelligence · Computer Science 2024-05-28 Miguel Angel Mendez-Lucero , Enrique Bojorquez Gallardo , Vaishak Belle

Neural networks excel across a wide range of tasks, yet remain black boxes. In particular, how their internal representations are shaped by the complexity of the input data and the problems they solve remains obscure. In this work, we…

Machine Learning · Computer Science 2026-05-12 Robert Jankowski , Filippo Radicchi , M. Ángeles Serrano , Marián Boguñá , Santo Fortunato

We present a generalisation of Rosenblatt's traditional perceptron learning algorithm to the class of proximal activation functions and demonstrate how this generalisation can be interpreted as an incremental gradient method applied to a…

Machine Learning · Computer Science 2020-12-08 Xiaoyu Wang , Martin Benning

We introduce an information-theoretic framework that views learning as universal prediction under log loss, characterized through regret bounds. Central to the framework is an effective notion of architecture-based model complexity, defined…

Machine Learning · Computer Science 2025-11-04 Meir Feder , Ruediger Urbanke , Yaniv Fogel

Classical analyses of gradient descent (GD) define a stability threshold based on the largest eigenvalue of the loss Hessian, often termed sharpness. When the learning rate lies below this threshold, training is stable and the loss…

Machine Learning · Computer Science 2025-11-18 Lawrence Wang , Stephen J. Roberts

Neural networks trained with standard objectives exhibit behaviors characteristic of probabilistic inference: soft clustering, prototype specialization, and Bayesian uncertainty tracking. These phenomena appear across architectures -- in…

Machine Learning · Computer Science 2026-01-01 Alan Oursland

The underspecification of most machine learning pipelines means that we cannot rely solely on validation performance to assess the robustness of deep learning systems to naturally occurring distribution shifts. Instead, making sure that a…

We define the local complexity of a neural network with continuous piecewise linear activations as a measure of the density of linear regions over an input data distribution. We show theoretically that ReLU networks that learn…

Machine Learning · Computer Science 2025-07-15 Niket Patel , Guido Montufar

Current methods to interpret deep learning models by generating saliency maps generally rely on two key assumptions. First, they use first-order approximations of the loss function neglecting higher-order terms such as the loss curvatures.…

Machine Learning · Computer Science 2019-06-03 Sahil Singla , Eric Wallace , Shi Feng , Soheil Feizi

Recent literature generalization in deep learning has examined the relationship between the curvature of the loss function at minima and generalization, mainly in the context of overparameterized neural networks. A key observation is that…

Machine Learning · Computer Science 2025-10-01 Neta Shoham , Liron Mor-Yosef , Haim Avron

Meta-learning that uses implicit gradient have provided an exciting alternative to standard techniques which depend on the trajectory of the inner loop training. Implicit meta-learning (IML), however, require computing $2^{nd}$ order…

Machine Learning · Computer Science 2023-10-31 Fady Rezk

Flat minima are strongly associated with improved generalisation in deep neural networks. However, this connection has proven nuanced in recent studies, with both theoretical counterexamples and empirical exceptions emerging in the…

Machine Learning · Computer Science 2026-04-16 Israel Mason-Williams , Gabryel Mason-Williams , Helen Yannakoudakis

Although neural networks are routinely and successfully trained in practice using simple gradient-based methods, most existing theoretical results are negative, showing that learning such networks is difficult, in a worst-case sense over…

Machine Learning · Computer Science 2017-03-13 Ohad Shamir