optimization

Published on : 2026-06-06 Reading time : 6 min Tags : #python #async #performance #optimization The Problem: Fake Async The supabase-async library claimed to be async but actually wrapped synchronous calls with ThreadPoolExecutor: # ❌ Fake async (old code) class SupabaseAsync : def __init__ ( self ): self . _executor = ThreadPoolExecutor ( max_workers = 3 ) async def select ( self , table : str )…

Application deadline: Tuesday, June 30, 2026 Employer web page: https://huberfe.github.io/jobs Job type: PhD PhD student (4 years). We are seeking a motivated PhD student that is interested to work on quantum information and mathematical optimization in the group of Felix Huber . Topics they can expect to work on are: approximation algorithms in QIT (e.g. ground state energies) symmetry- and size…

Matrix problems often seem tricky because multiple rows and columns move at the same time. In this problem, we need to rotate an n × n matrix by 90 degrees clockwise without using any extra matrix. The key insight is that we don't need to perform the rotation directly. Instead, we can break it down into two simple operations that are easy to perform in-place. Problem Statement Given an n x n matr…

Fine-tuning for domain-specific tasks means improving performance in one area without degrading the model’s general capabilities, and getting that balance right is harder than it looks. This post walks through how to navigate that balance, from selecting the right customization strategy for your data and task, to configuring the training parameters that most influence outcomes, like learning rate…

Let's chat about performance, optimization, and debugging a PostgreSQL database. Can you read EXPLAIN ANALYZE output fluently? Do you know how to find which tenant is causing a slow query, how to use pg_stat_statements , how to spot lock contention? If you have a hard time answering those, don't worry. Today we're going to cover all of them and practice how to get better at it. The goal here: you…

Late physicist turned issue of when to stop searching for a better place to eat into mathematical problem When it comes to exploring a new city, it can be tricky to know when to stop searching for a different restaurant to try every night, or to visit the first place you love on repeat. Now researchers have found that the late physicist and Nobel laureate Richard Feynman devised a mathematical eq…

Physicist Richard Feynman turned a lunch dilemma into a math problem. Researchers finally cracked his notes and found people approximate his solution on their own.

Mathematics has always been the language of deep structure. It gives us a way to describe patterns, reason about complexity, and understand what is possible. It is also one of the foundations of modern computing. Every processor instruction, every algorithm, and every computational breakthrough ultimately rests on mathematical ideas about how information can be represented, transformed, optimized…

Let $$ \ell_1^{(1)} = \left\{ x=(x_i)_{i\ge1} : x_i\ge 0,\; \sum_{i=1}^{\infty} x_i =1 \right\}. $$ Equip $\ell_1^{(1)}$ with the metric induced by the $\ell^1$ norm: $$ d(x,y) = \|x-y\|_1 = ...

Scientific Reports, Published online: 01 June 2026; doi:10.1038/s41598-026-55326-x Multi-Strategy optimisation algorithm for multi-UAV path planning in complex mountainous environments

I am a second year PhD student in mathematics and I have been working on a problem for a couple years. I published some papers and gave conference talks on partial results. I feel I am very close to ...

Scientific Reports, Published online: 30 May 2026; doi:10.1038/s41598-026-54201-z A state-adaptive booby optimization algorithm for engineering design and medical data applications

There is a common measure of wiggliness of a curve which is the integral of the square of its second derivative. I don't know if it has a name, or any formal basis, although perhaps it is related to ...

According to a mathematical model of how people weigh up different outcomes, the optimal strategy is to be ambitious, but not overly so

A step-by-step journey from calculus-based optimization to Stochastic Gradient Descent The post Why Gradient Descent Became Stochastic appeared first on Towards Data Science .

A mathematician found the most efficient way to fold paper into a doughnutlike shape.

Generalized low-rank matrix recovery is a fundamental statistical tool that provides a unifying framework for important data science problems such as Euclidean distance geometry, robust principal component analysis, and matrix completion. It arises in applications such as sensor localization, molecular conformation, ultrasound imaging, video processing, and hyperspectral imaging. In these applica…

Model created by researchers shows better outcomes are often more likely when people are not too ambitious It is the end of an idiom for motivational speakers. Instead of shooting for the moon when pursuing life’s goals, researchers say people should be advised to aim a little lower if they want the best outcome. The tip may lack the punch of uncompromising drive, but aiming for merely above aver…

In this work, a Coupled Radial Basis Function (CRBF) and cover scheme method is presented to solve nonlinear optimal control problems (OCPs) in space flight applications. Since OCPs are often very complex, numerical methods are used to solve them. Direct methods transcribe the OCP into a nonlinear programming problem (NLP) by discretizing the continuous-time problem.  Indirect methods use Pontrya…

Modern AI systems are increasingly optimized for higher mathematical precision and stricter alignment. While this reduces local prediction error, it often leads to a paradoxical outcome: existing users feel the system has become colder, more rigid, and less usable. This paper argues that this phenomenon stems from the trap of excessive precision, which dramatically increases boundary friction (F)…