Life & fun · alternatives · 2026

24 alternatives to Per-instance TSP Solver with No Pre-training (1.66% gap on d1291)

OP here. Most Deep Learning approaches for TSP rely on pre-training with large-scale datasets. I wanted to see if a solver could learn "on the fly" for a specific instance without any priors from other problems. I built a solver using PPO that learns from scratch per instance. It achieved a 1.66% gap on TSPLIB d1291 in about 5.6 hours on a single A100. The Core Idea: My hypothesis was that while optimal solutions are mostly composed of 'minimum edges' (nearest neighbors), the actual difficulty comes from a small number of 'exception edges' outside of that local scope. Instead of…

This is a traveling salesman problem solver that learns independently for each problem instance using reinforcement learning, without relying on pre-trained models from other datasets. Built with PPO, it achieved a… Below are 24 products that do a similar job, ranked by how close each is in meaning and then by launch-day votes.

1
Watch a neural net learn to play Snake▲205
In browser PPO training demo, made possible by tinygrad: TinyJit -> WebGPU kernels. Requires WebGPU.
May 2026 · ppo.gradexp.xyz · its alternatives →
2
How I topped the HuggingFace open LLM leaderboard on two gaming GPUs▲495
I found that duplicating a specific block of 7 middle layers in Qwen2-72B, without modifying any weights, improved performance across all Open LLM Leaderboard benchmarks and took #1. As of 2026, the top 4 models on that leaderboard are still descendants. The weird finding: single-layer duplication does nothing. Too few layers, nothing. Too many, it gets worse. Only circuit-sized blocks of ~7 layers work. This suggests pretraining carves out discrete functional circuits in the layer stack that only work when preserved whole. The whole thing was developed on 2x RTX 4090s in my basement. I'm…
Mar 2026 · dnhkng.github.io · its alternatives →
3
Interactive visual solver for the traveling salesman problem▲113
Interactive solver for the traveling salesman problem to visualize different algorithms. Includes various Heuristic and Exhaustive algorithms.
2019 · tspvis.com · its alternatives →
4BP
Beating Pokemon Red with RL and <10M Parameters▲183
Hi everyone! After spending hundreds of hours, we're excited to finally share our progress in developing a reinforcement learning system to beat Pokémon Red. Our system successfully completes the game using a policy under 10M parameters, PPO, and a few novel techniques. With the release of Claude Plays Pokémon, now feels like the perfect time to showcase our work. We'd love to get feedback!
2025 · drubinstein.github.io · its alternatives →
5
Fast Deep Reinforcement Learning Course▲148
I worked on this applied Deep Reinforcement Learning course for the better part of 2021. I made a Datacamp course [0] before, and this served as my inspiration to make an applied Deep RL series. Normally, Deep RL courses teach a lot of mathematically involved theory. You get the practical applications near the end (if at all). I have tried to turn that on its head. In the top-down approach, you learn practical skills first, then go deeper later. This is much more fun. This course (the first in a planned multi-part series) shows how to use the Deep Reinforcement Learning framework RLlib to…
2022 · courses.dibya.online · its alternatives →
6
Boldly go where Gradient Descent has never gone before with DiscoGrad▲232
Trying to do gradient descent using automatic differentiation over branchy programs? Or to combine them with neural networks for end-to-end training? Then this might be interesting to you. We develped DiscoGrad, a tool for automatic differentiation through C++ programs involving input-dependent control flow (e.g., "if (f(x) < c) { ... }", differentiating wrt. x) and randomness. Our initial motivation was to enable the use of gradient descent with simulations, which often rely heavily on such discrete branching. The latter makes plain autodiff mostly useless, since it can only account for the…
2024 · github.com · its alternatives →
7IB
I built a toy TPU that can do inference and training on the XOR problem▲134
We wanted to do something very challenging to prove to ourselves that we can do anything we put our mind to. The reasoning for why we chose to build a toy TPU specifically is fairly simple: - Building a chip for ML workloads seemed cool - There was no well-documented open source repo for an ML accelerator that performed both inference and training None of us have real professional experience in hardware design, which, in a way, made the TPU even more appealing since we weren't able to estimate exactly how difficult it would be. As we worked on the initial stages of this project, we…
2025 · tinytpu.com · its alternatives →
8
Terminal-Bench-RL: Training long-horizon terminal agents with RL▲125
After training calculator agent via RL, I really wanted to go bigger! So I built RL infrastructure for training long-horizon terminal/coding agents that scales from 2x A100s to 32x H100s (~$1M worth of compute!) Without any training, my 32B agent hit #19 on Terminal-Bench leaderboard, beating Stanford's Terminus-Qwen3-235B-A22! With training... well, too expensive, but I bet the results would be good! *What I did*: - Created a Claude Code-inspired agent (system msg + tools) - Built Docker-isolated GRPO training where each rollout gets its own container - Developed a multi-agent…
2025 · github.com · its alternatives →
9
Iteratively Building Virtual Creatures in Minecraft▲49
"Creatures" is a stretch given that the environment is Minecraft, but the idea is simple: iteratively add blocks conditionally (tensor convolution) on the current environment (blocks) to maximize some reward. In this case I use PPO RL to train creatures to touch a glowstone block but you can adapt it to use any algorithm and reward (easily, as it uses the Ray framework). What I like about this work: iteratively finding solutions has a long and colorful history of doing things well: gradient boosting, ResNets, Stable Diffusion, etc. We're after some end optimal state and usually try to get…
2024 · github.com · its alternatives →
10
I RL-trained an agent that trains models with RL (for ~$1.3k)▲107
RL-training an AI agent to RL-train AI agents. Contribute to Danau5tin/ai-trains-ai development by creating an account on GitHub.
Jul 2026 · github.com · its alternatives →
11
Create your own finetuned AI model using Google Sheets▲137
Hello HN, We built Promptrepo to make finetuning accessible to product teams — not just ML engineers. Last week, OpenAI’s CPO shared how they use fine-tuning for everything from customer support to deep research, and called it the future for serious AI teams. Yet most teams I know still rely on prompting, because fine-tuning is too technical, while the people who have the training data (product managers and domain experts) are often non-technical. With Promptrepo, they can now: - Add training examples in Google Sheets - Click a button to train - Deploy and test instantly - Use OpenAI,…
2025 · promptrepo.com · its alternatives →
12
A perceptron with stochastic gradient descent in Go▲94
A single / multi layer / recurrent neural network written in Golang. - made2591/go-perceptron-go
2017 · github.com · its alternatives →
13
A TSP game I wanted for 10 years - built in 4 hours▲12
For a decade, I’ve wanted to build a simple game that shows why the Traveling Salesman Problem is fun to work on. Something I could pull up at a party when someone asks “so what does your company actually do?” Something my kids could play and understand why dad stares at vehicle ro
Dec 2025 · graphhopper.com · its alternatives →
14
A radically more scalable alternative to Bayesian Optimization▲96
Experimental Global Optimization Algorithm. Contribute to chrisstroemel/Simple development by creating an account on GitHub.
2018 · github.com · its alternatives →
15
Next-Gen AI Training: LLM-RLHF-Tuning with PPO and DPO▲30
Comprehensive toolkit for Reinforcement Learning from Human Feedback (RLHF) training, featuring instruction fine-tuning, reward model training, and support for PPO and DPO algorithms with various configurations for the Transformer 5 models including Qwen 3.0 - raghavc/LLM-RLHF-Tuning-with-PPO-and-DP
2024 · github.com · its alternatives →
16
Multi-GPU Reinforcement Learning in Tensorflow for OpenAI Gym▲58
Multi-GPU reinforcement learning using Deep Q-Network in TensorFlow for OpenAI Gym - vishar0/dist-dqn
2016 · github.com · its alternatives →
17
Minimal LLM Post-Training Experiments on an 8GB GPU (SFT, DPO, GRPO)▲21
Minimal, readable LLM post-training experiments on one 8GB GPU. Measures forgetting, seed variance, and RL emergence. - pochenai/nano-llm-posttraining
Aug 2026 · github.com · its alternatives →
18
Python 3 TSP solver based on LKH (cross platform)▲57
elkai is a Python library for approx. solving travelling salesman problems (TSP) based on LKH 3 - fikisipi/elkai
2019 · github.com · its alternatives →
19IC
I created a Neural Network from scratch, in scratch▲7
I spent the past week implementing a 1 Layer Neural Net and training it on MNIST within the visual scripting language provided by scratch.mit.edu. It was tedious, but ultimately not too difficult. The code runs incredibly slowly, so much so that 64 samples of MNIST takes 5+ hours to train on my machine. There were a lot of little mini challenges that were fun to overcome (implementing softmax was very tricky). If you're interested, I encourage you to try and improve on it! More details in the linked blog post.
2024 · bell-boy.github.io · its alternatives →
20AB
A batched, concurrent wrapper for OpenAI gym library▲12
Hello HN, new user here, so please let me know if I break some rules. Currently I've been working on training reinforcement learning agents, and OpenAI gym, while is great, runs only one agent at a time. Hence I decided to extend it. I built a wrapper around OpenAI gym, such that it now runs several environments concurrently. All while (mostly) having the same call signature as OpenAI gym. And it is published to PyPI for anyone interested. For more details, please visit: https://github.com/Chimpan-Z/agymc Feedback really appreciated! Have a good day everyone!
2020 · its alternatives →
21
I built a deep learning engine from scratch in Python▲30
I’ve spent the last few months building a deep learning engine completely from scratch in Python (using only math and random). What started as a basic linear algebra calculator project grew into a symbolic tensor system with autodiff, custom matrix ops, attention mechanisms, LayerNorm, GELU, and even a text generation demo trained on the Brown corpus. I'm still an undergrad, so my main goal is to deeply understand how deep learning actually works under the hood - gradients, attention, backpropagation, optimizers - by building it step-by-step with full visibility into everything, and without…
2025 · github.com · its alternatives →
22RA
Relari – Auto Prompt Optimizer as Lightweight Alternative to Finetuning▲32
Hi HN, we are the founders of Relari (https://www.relari.ai). We launched our LLM evaluation stack on HN a few months ago (https://news.ycombinator.com/item?id=39641105), which is now used in production by AI teams at companies like Vanta and PwC. We have since expanded to directly optimizing parts of an LLM pipeline using a data-driven approach. In particular, we see a lot of potential in the Auto Prompt Optimization—which could be an attractive alternative to fine-tuning in many cases—to use data to align LLMs for domain-specific tasks. Here’s a demo video:…
2024 · its alternatives →
23W1
Watch 14-Byte AI "brains" attempt to solve a 2D maze (Its hard)▲25
Hey HackerNews, I built this project over the last few weeks as a palette cleanser from a failed game launch. I wanted to learn a bit about AI/Neural-Networks and naively thought I could build a tiny maze-solving AI in a weekend with a 100% solve rate. Well - I couldn't, but I got pretty close. 14 Bytes total model size, and a 96.5% solve rate on unseen mazes. Trained across 46 phases experimenting with different ideas to improve the model (better performance, smaller size). Its quite fun to watch the model attempt to solve the maze, when they fail its usually due to getting stuck in a…
Jul 2026 · con-dog.github.io · its alternatives →
24
I'm 15 and built a neural network from scratch in C++,just math▲8
I’m 15 and self-taught. I'm learning ML from scratch because I want to really understand how things work. I’m not into frameworks. I prefer math, logic, and C++. I implemented a basic MLP that supports different activation and loss functions. It was trained via mini-batch gradient descent. I wrote it from scratch, using no external libraries except Eigen (for linear algebra). I learned how a Neural Network learns (all the math) -- how the forward pass works, and how learning via backpropagation works. How to convert all that math into code. I’ll write a blog soon explaining how MLPs work in…
2025 · github.com · its alternatives →

Also compare

Ranked by how close each launch is in meaning, then by votes. Prices were read from each product’s own site when checked and can change. Refine with your own description →