Deep neural networks constructed from linear maps and positively homogeneous nonlinearities (e.g., ReLU) possess a fundamental gauge symmetry: the network function is invariant to node-wise diagonal rescalings. However, standard gradient descent is not equivariant to this symmetry, causing optimization trajectories to depend heavily on arbitrary parameterizations. Prior work has proposed rescaling-invariant optimization schemes for positively homogeneous networks (e.g., path-based or path-space updates). Our contribution is complementary: we formulate the invariance requirement at the level of the backward adjoint/optimization […]

Ver mais

arXiv:2412.19217v4 Announce Type: replace-cross Abstract: The rapid expansion of citizen science initiatives has led to a significant growth of biodiversity databases, and particularly presence-only (PO) observations. PO data are invaluable for understanding species distributions and their dynamics, but their use in a Species Distribution Model (SDM) is curtailed by sampling biases and the lack of information on absences. Poisson point processes are widely used for SDMs, with Maxent being one of the most popular methods. Maxent maximises the […]

Ver mais

arXiv:2601.19947v1 Announce Type: new Abstract: Learning from Noisy Labels (LNL) presents a fundamental challenge in deep learning, as real-world datasets often contain erroneous or corrupted annotations, textit{e.g.}, data crawled from Web. Current research focuses on sophisticated label correction mechanisms. In contrast, this paper adopts a novel perspective by establishing a theoretical analysis the relationship between flatness of the loss landscape and the presence of label noise. In this paper, we theoretically demonstrate that carefully simulated label noise synergistically […]

Ver mais

This work presents a real-time default-detection model integrating streaming behavioral signals, including app usage dynamics, repayment timing, transaction irregularities, and short-term income proxies. The model is built on 2.7 million active loan accounts with second-level event streams. A streaming-enabled ensemble classifier (online gradient boosting + incremental random forest) is deployed with a sliding window of 7 days. The model predicts impending 30-day delinquency with an ROC-AUC of 0.89 and reduces detection delay by 9.2 days on average. Incorporating […]

Ver mais

arXiv:2602.18014v1 Announce Type: cross Abstract: Repetitive motion tasks are common in robotics, but performance can degrade over time due to environmental changes and robot wear and tear. Iterative learning control (ILC) improves performance by using information from previous iterations to compensate for expected errors in future iterations. This work incorporates the use of Quasi-Periodic Gaussian Processes (QPGPs) into a predictive ILC framework to model and forecast disturbances and drift across iterations. Using a recent structural equation formulation of […]

Ver mais

The code needed configuration values. AI GENERATED # config.pyDATABASE_URL = “postgresql://localhost/mydb”API_KEY = “sk_live_abc123″DEBUG = True# app.pyimport configdef connect_database(): return psycopg2.connect(config.DATABASE_URL)def call_api(endpoint): return requests.get(f”https://api.example.com/{endpoint}”, headers={‘X-API-Key’: config.API_KEY}) Clean. Simple. Every module could access configuration through import config. No passing parameters everywhere. Convenient. Then the bug reports started. “The app works fine on my machine but not in production.” “Tests pass locally but fail in CI.” “Feature works when run alone but breaks when run after other tests.” I couldn’t reproduce any of […]

Ver mais

So I am using a R(2+1)D with kinetics 400 weights to train a classifier on two sets of videos. The problem is that one of the two classes has all videos of the same resolution and fps, forcing the model to learn those features instead of actually learning pixel changes over time, like R(2+1)D is supposed to. On the other class, there is diversity and equivalent representation across resolutions, which makes the model totally unusable without any preprocessing. […]

Ver mais

How are you, hacker? 🪐Want to know what’s trending right now?: The Techbeat by HackerNoon has got you covered with fresh content from our trending stories of the day! Set email preference here. ## Your AI Model Isn’t Broken. Your Data Is By @melissaindia [ 7 Min read ] Your AI model isn’t failing; your data is. Learn how clean, verified data improves model accuracy and how easy it is to fix with APIs. Read More. The Hidden […]

Ver mais

I recently made the code for our recently published project, Vejde, public. It was originally built to handle variably sized inputs in automated network intrusion response, but we made and did an evaluation of a generic version in order to allow it to be used for other problem domains as well. Since I sometimes see people struggling with problems that this might be useful for in this subreddit, I thought it might be prudent to also inform about […]

Ver mais

Cardiac blood flow patterns contain rich information about disease severity and clinical interventions, yet current imaging and computational methods fail to capture underlying relational structures of coherent flow features. We propose a physics-informed, latent relational framework to model cardiac vortices as interacting nodes in a graph. Our model combines a neural relational inference architecture with physics-inspired interaction energy and birth-death dynamics, yielding a latent graph sensitive to disease severity and intervention level. We first apply this to computational […]

Ver mais