Konstantin Pandl

Performed some memory profiling with ~2.5 hours of 25 TPS [here](https://github.com/AleoHQ/stress-observability/issues/115#issuecomment-2072450510). After the test, snarkOS maximum memory usage was reduced by 8% and average memory usage was reduced by 4.4%...

**Milestone 1 results** The experiments are conducted in the 6 Jupyter notebooks in zkml-research/MNIST_preprocessing. The last notebook (6_presentation) has the most promising results in terms of accuracy and expected computational...

**Milestone 2 results** For milestone 2, I mainly used the mlp1 Leo program, and variations thereof (e.g., by commenting out certain sections or changing few lines of code and measuring...

**Milestone 3 results** For milestone 3, I implemented MLP neural networks in the transpiler, and released a new transpiler version on PyPi. I also updated the decision tree MNIST example...

Generally looks good to me. Can you explain the setting of the const variables (`MIN_CONNECTION_INTERVAL_IN_SECS` and `IP_BAN_TIME_IN_SECS`) or do you know recommended values from other protocols? I understand they can...

Thanks for the feedback! I've updated the code. The `node` test ran successfully locally. The `node-bft` test did not, but that was likely due to constraint local resources, not an...

@zosorock thanks, I've merged the latest `staging` into this branch, there should be no conflicts now

Based on a flamegraph analysis, the most time-consuming operation within the `check_next_block` function is `compute_finalize_root`. While there is potential to further optimize by caching, it seems worth investigating how we...

Tested it on a 5 validator network, with 4 instance doing nested deployments, 4 instances doing nested executions, and a program probe instance running every 500 ms (total ~100 program...