7 ms·
Ask HN: ML Papers to Implement
What ML papers do you think someone who has a some experience - but would still consider themselves relatively inexperienced in ML, could implement?
ideally, a list of papers that could take 2-5 hours each and a few hundred lines of code?
- m-watson 4y agoThis is not exactly what you are looking for but you should browse Papers with Code: https://paperswithcode.com/ https://paperswithcode.com/
- Buttons840 4y agoSoft Actor Critic, a reinforcement learning algorithm.
- Heidaradar 4y agothanks!
- ausbah 4y agoyou could maybe write the whole thing in a few hours, but debugging what you wrote to recreate prior results will probably take much longer depending on choice of problem
- Heidaradar 4y agowhat paper is this referring to, sorry?
- ausbah 4y agosorry should've been more specific, but ML is general. from what i've seen it's not hard to reimplement the pseudo-code in any ML paper. it's just that it gets tricky when you actually try to utilize the code you've written, usually in trying to recreate performance results in the implementing paper. it's very common for authors to leave out / downplay the role of tricks or implementation details that greatly contributed to the performance of the model, in addition to just how finicky machine learning is in general
- Heidaradar 4y agoi see, that makes sense. thanks for the info.
- polygamous_bat 4y agoI would recommend diffusion: try starting with Lilian Weng's blog post and writing up the process for yourself. For all it's abilities, the code for DDPM is surprisingly simple.
- naillo 4y agoThis is a great from-scratch (non bloaty) tutorial on this imo https://colab.research.google.com/github/huggingface/diffusion-models-class/blob/main/unit1/02_diffusion_models_from_scratch.ipynb https://colab.research.google.com/github/huggingface/diffusi...
- phonebucket 4y agoA lot depends on what you're interested in. Some papers that are runnable on a laptop CPU (so long as you stick to small image sizes/tasks): 1) Generative Adversarial Networks (https://arxiv.org/abs/1406.2661 https://arxiv.org/abs/1406.2661). Good practice to have a custom training loops, different optimisers and networks etc. 2) Neural Style Transfer (https://arxiv.org/abs/1508.06576 https://arxiv.org/abs/1508.06576). Nice to be able to manipulate pretrained networks and intercept intermediate layers. 3) Deep Image Prior (https://arxiv.org/abs/1711.10925 https://arxiv.org/abs/1711.10925). Nice low-data exercise in building out an autoencoder. 4) Physics Informed Neural Networks (https://arxiv.org/abs/1711.10561 https://arxiv.org/abs/1711.10561). If you're interested scientific applications, this might be fun. It's good exercise in calculating higher order derivatives of neural networks and using these in loss functions. 5) Vanilla Policy Gradient (https://arxiv.org/abs/1604.06778 https://arxiv.org/abs/1604.06778) is the easiest reinforcement learning algorithm to implement and can be used as a black-box optimiser in a lot of settings. 6) Deep Q Learning (https://arxiv.org/abs/1312.5602 https://arxiv.org/abs/1312.5602) is also not too hard to implement and was the first time I had heard about DeepMind, as well as being a foundational deep reinforcement learning paper . Open AI gym (https://github.com/openai/gym https://github.com/openai/gym) would help get started with the latter two.
- OgAstorga 4y agothank you, this is a great compilation
- Heidaradar 4y agothis is exactly what I needed, thanks!
- Artgor 4y agoYou could start with Word2Vec or GloVe, for example. Another option is to to with CV papers and start with Alexnet or first ResNet.
- Heidaradar 4y agothanks!
- imranq 4y agoHighly recommend this resource for RL https://spinningup.openai.com/en/latest/ https://spinningup.openai.com/en/latest/
- Heidaradar 4y agogreat, thanks!
- la_fayette 4y agoI have implemented YOLO v1 and train/tested it on synthetic images with geometric forms. Implementing the loss function thought me a lot on how backpropagation really works. I used keras/tf.
- Heidaradar 4y agohow long did that take
- carom 4y agoAndrej Karpathy is currently releasing videos for a course [1] that goes from zero to GPT. 1. https://karpathy.ai/zero-to-hero.html https://karpathy.ai/zero-to-hero.html
- Heidaradar 4y agothanks!
- manimino 4y agoI enjoyed doing this through the Coursera deep learning specialization: https://www.coursera.org/specializations/deep-learning https://www.coursera.org/specializations/deep-learning The lectures take you through each major paper, then you implement the paper in the homework. Much faster than reading the paper yourself.
- Heidaradar 4y agohmm, I found that I didn't really like this course but I'll look through it again
- osaariki 4y agoI'd love for someone to do a good quality PyTorch enabled implementation of Sampled AlphaZero/MuZero [1]. RLLib has an AlphaZero, but it doesn't have the parallelized MCTS you really want to have and the "Sampled" part is another twist to it. It does implement a single player variant though, which I needed. This would be amazing for applying MCTS based RL to various hard combinatorial optimization problems. Case in point, AlphaTensor uses their internal implementation of Sampled AlphaZero. An initial implementation might be doable in 5 hours for someone competent and familiar with RLLib's APIs, but could take much longer to really polish. [1]: https://arxiv.org/abs/2104.06303 https://arxiv.org/abs/2104.06303
- Heidaradar 4y agoI'll definitely take a look!
- tdekken 4y agoGreat question! I seem to be in a similar situation as an experienced software engineer who has jumped into the deep end of ML. It seems most resources either abstract away too much detail or too little. For example, building a toy example that just calls gensim.word2vec doesn't help me transfer that knowledge to other use cases. Yet on the other extreme, most research papers are impenetrable walls of math that obscure the forest for the trees. Thus far, I would also recommend Andrej Karpathy's Zero to Hero course (https://karpathy.ai/zero-to-hero.html https://karpathy.ai/zero-to-hero.html). He assumes a high level of programming knowledge but demystifies the ML side. -- P.S. If anyone is, by chance, interested in helping chip away at the literacy crisis (e.g., 40% of US 4th graders can't read even at a basic level), I would love to find a collaborator for evaluating the practical application of results from the ML fields of cognitive modeling and machine teaching. These seemingly simple ML models offer powerful insight into the neural basis for learning but are explained in the most obtuse ways.
- Heidaradar 4y agogreat, thanks! also for your PS, can you give a little more detail? What's your end result, what have you done so far etc
- tdekken 4y agoI would love to! My main goal is to use cognitive modeling to evaluate the efficacy of interventions and inform the personalized "minimum effective dose" for a particular learner. Academically, this is well-trodden territory [0-2] but these results haven't found there way into practice. This is critically important because we know that ~30% of children will learn to read regardless of method, ~50% require explicit, systematic instruction, ~15% require prolonged explicit and systematic instruction, and up to 6% have severe cognitive impairments that make acquiring reading skills extremely difficult [3]. Yet, how much is enough? To make this more concrete, imagine you are learning a foreign language with Duolingo. How much effort per day is necessary to achieve that? Many people have long streaks and are no closer to fluency (I learned nearly nothing despite a 400 day streak). Similarly, many reading interventions are once-a-week and, predictably, don't meaningfully affect the learning outcomes for those students. BTW, this ML portion is part of a much larger effort (e.g., our team is a Phase II finalist in the Learning Engineering Tools Competition). If anyone is interested in collaborating, please feel free to reach out to me. [0] Phonology, reading acquisition, and dyslexia: insights from connectionist models (https://pubmed.ncbi.nlm.nih.gov/10467896/ https://pubmed.ncbi.nlm.nih.gov/10467896/) [1] Modeling the successes and failures of interventions for disabled readers. (https://www.researchgate.net/publication/243777699_Modeling_the_Successes_and_Failures_of_Interventions_for_Disabled_Readers https://www.researchgate.net/publication/243777699_Modeling_...) [2] Learning to Read through Machine Teaching (https://arxiv.org/abs/2006.16470 https://arxiv.org/abs/2006.16470) [3] Education Advisory Board. (2019). Narrowing the Third-grade Reading Gap: Embracing the Science of Reading, District Leadership Forum: Research briefing
- Mxbonn 4y agoNot sure if it's beginner friendly but I found implementing NeRF from scratch a good exercise. Especially since it reveals many details that are not immediately obvious from the paper.
- jahewson 4y agoSounds neat. Tell me more.
- kkaranth 4y agoI've been working on implementing this too! It's been fun trying to debug problems by figuring out how to reduce it to the "bare minimum" to repro. Ex: does it work if I disable positional encoding? Does it work if I have only one sample per ray on a single image dataset? Etc
- KhoomeiK 4y agoHey, feel free to reach out if you’d like to join an NLP project to gain more experience that I’m working on. Will provide mentorship and potentially coauthorship on the publication.
- freetheelephant 4y agoHey, I'd love to potentially join and help out on this project and learn more about NLP.
- KhoomeiK 4y agoSure, feel free to email me as well. Email is listed on the personal website linked in my profile.
- Heidaradar 4y agoyeah, i'd love to! how do you want me to message you?
- KhoomeiK 4y agoEmail is on my personal website linked in my profile
- Heidaradar 4y agosent! - (yesterday)
- zebloxxer 4y agoHey - I would love to help out with whatever needs doing. Happy to do grunt work. I've been deeply studying NLP for a few months now and a project is exactly what I need to help me move forward with it.
- KhoomeiK 4y agoSure, feel free to email me as well. Email is listed on the personal website linked in my profile.
- p1esk 4y agoIt's best to choose something you personally find interesting, for example, I'm interested in audio generation, so I'd pick some papers that describe a music/voice generation model or algorithm, but to you it might be something completely different. When you do decide on a paper, take a look at Phil Wang's implementation style: https://github.com/lucidrains?tab=repositories https://github.com/lucidrains?tab=repositories, he has hundreds of papers implemented. If you don't already have a GPU machine, you can rent 40GB A100 instance for $1.1/hr or 24GB A10 for $0.6/hr: https://lambdalabs.com/service/gpu-cloud https://lambdalabs.com/service/gpu-cloud.
- Heidaradar 4y agoit's hard for me right now to tell if something is simple or not to implement, but i totally agree! thanks for the links.
- badpun 4y ago2-5 hours for a few hundred lines of tricky math code sounds like way too little. Not to mention, having to read and understand the paper first. Depending on the difficulty of the paper and your level of skill in the field, I'd say implementing a paper should take 20-200 hours.
- Heidaradar 4y agoOh interesting, didn't know it'd take so long lol Any recommendations for papers you've implemented before?
- badpun 4y agoSorry, I mostly work on (hobbyist) 3d computer vision and not ML. BTW time of implementation also greatly depends on what you've implemented already. Most papers are a small derivation of some preexisting idea, so I've you've already implemented that idea, there isn't that much work to do on top of it - just modify your existing code. But, if you're just starting with some area, getting up to that point will take time.
- Heidaradar 4y agoOh i see, thanks for the info.
- paulmorio 4y agoI have done this a few times now. Alone (e.g. https://github.com/paulmorio/geo2dr https://github.com/paulmorio/geo2dr) and in collaboration with others (e.g. https://github.com/benedekrozemberczki/pytorch_geometric_temporal https://github.com/benedekrozemberczki/pytorch_geometric_tem...) primarily as a way to learn about the methods I was interested in from a research perspective whilst improving my skills in software engineering. I am still learning. Starting out I would recommend implementing fundamental building blocks within whatever 'subculture' of ML you are interested in whether that be DL, kernel methods, probabilistic models, etc. Let's say you are interested in deep learning methods (as that's something I could at least speak more confidently about). In that case build yourself an MLP layer, then an RNN layer, then a GNN layer, then a CNN layer, and an attention layer along with some full models with those layers on some case studies exhibiting different data modalities (images, graphs, signals). This should give you a feel for the assumptions driving the inductive biases in each layer and what motivates their existence (vs. an MLP). It also gives you the all the building blocks you can then extend to build every other DL layer+model out there. Another reason is that these fundamental building blocks have been implemented many times so you have a reference to look to when you get stuck. On that note: here are some fun GNN papers to implement in order of increasing difficulty (try building using vanilla PyTorch/Jax instead of PyG). - SGC (from https://arxiv.org/abs/1902.07153 https://arxiv.org/abs/1902.07153) - GCN (from https://arxiv.org/abs/1609.02907 https://arxiv.org/abs/1609.02907) - GAT (from https://arxiv.org/abs/1710.10903 https://arxiv.org/abs/1710.10903) After building the basic building blocks these should each take about 2-5 hours (reading paper + implementation). Probably quicker at the end with all this practice. Good luck and remember to have fun!
- Heidaradar 4y agoquick question: when you say write your own MLP, RNN etc, is that without using PyTorch, I'm assuming so? i.e., i'm guessing you want me to write my own NN library that handles all of this stuff, including backprop etc
- Heidaradar 4y agoanother question: Do you have a good resource for learning more about GNNs, I'm currently looking at the Stanford course, is that good enough? any other courses/books/ just something else that you think would be more useful?
- biotechbio 4y agoJust finished assignment 2 of cs224n[1], which has you derive gradients and implement word2vec. I thought it was a pretty good exercise. You could read the glove paper and try implementing that as well. Knowing how to step through backpropagation in a neural network gets you pretty far in conceptual understanding of a lot of architectures. Imo there’s no substitute for writing out the gradients by hand to make sure you get what’s going on, if only in a toy example. [1] https://web.stanford.edu/class/cs224n/ https://web.stanford.edu/class/cs224n/
- Heidaradar 4y agothanks!
- loveparade 4y ago> ideally, a list of papers that could take 2-5 hours each and a few hundred lines of code? I think you are severely underestimating the time required, unless you are quite experienced, know exactly what to look for, or the paper is just a slight variation on previous work that you are already familiar with. Even seasoned researchers can easily spend 30+ hours on trying to reproduce a paper, because papers almost never contain all the details that went into the experiments. You are left with a lot of fiddling and iteration. Of course, if you only care about roughly reproducing what the authors did, and don't care about getting the same results, the time can be much shorter. If the code is available that's even better, but looking at it is cheating since wrestling with issues yourself is a big part of the learning process. A few people here mentioned Andrej's lectures, and I also think they are amazing, but they are not a replacement for getting stuck and solving problems yourself. You can easily watch these lectures and think "I get it!" because everything is so well explained, but you'll probably still be stuck when you run into your own problems trying to reproduce papers from scratch. There's no replacement for the experience you gain by struggling :) It's like watching a math lecture and thinking you get it, but then getting stuck at the exercise problems. The real learning happens when you force yourself to struggle through the exercises.
- Heidaradar 4y agoI completely agree (that struggling is the most important part, and also about andrej's lectures - they almost spoon feed you imo), but I also didn't know papers took so long. Any particular papers that you remember implementing yourself that you would reccomend?
- loveparade 4y agoI believe which paper you implement matters less than you think. Most papers take some kind of well-known base model, make a few tweaks, and then run experiments. To implement it you need to follow the references and start with the base model, and 99% of all papers within some field will eventually lead you back to implementing the same base model, so it really doesn't matter where you start. The overlap is huge.
- pizza 4y agoBiggest productivity boosters for me: - devdocs.io for pytorch - conda for packaging w CUDA - einops - tensorboard - huggingface datasets Interesting models/structures: - resnet - unet - transformers - convnext - vision transformers - ddpm - novel optimizers - generative flow nets - “your classifier is secretly an energy-based model and yiu should treat it like one” paper - self-supervision - distance metric learning Places where you can read implementations: - lucidrains’ github - timm computer vision models library - fastai - labml (annotated quite nicely) Biggest foreseeable headaches: - not suuper easy to do test-driven development - data normalization (floating point error, not using eg batchnorm) - sensitivity of model performance to (hyper)params (layer sizes, learning rates, optimizer, etc) - impatience - lack of data I’d also recommend watching Mark Saroufim live code in PyTorch, on YouTube. My 2 cents, you can only get really fast as well as good at this with a lot of experience. A lot of rules-of-thumb have to come together just right for the whole system to work.
- gschoeni 4y agoEarly days, but trying to solve the lack of data problem with a project called "Oxen". At it's core it's a data version control system. Built from the ground up to handle the size and scale of some of these ML datasets (unlike git-lfs or DVC which I have found to be relatively slow and hard to work with). Also building out a web hub similar to GitHub to collaborate on the data with a nice UI. Would love any feedback on the project as it grows! Here's the github repo: https://github.com/Oxen-AI/oxen-release#-oxen https://github.com/Oxen-AI/oxen-release#-oxen
- Heidaradar 4y agothanks!