Thursday, December 21, 2017

What is ablation study in machine learning

We often come across 'ablation study' in machine learning papers, for example, in this paper with the original R-CNN, it has a section of ablation studies. But what does this means?

Well, we know that when we build a model, we usually have different components of the model. If we remove some component of the model, what's the effect on the model? This is a very coarse definition of ablation study - we want to see the contributions of some proposed components in the model by comparing the model including this component with that without this component.

In the above paper, in order to see the effect of fine-tuning of the CNN, the authors analyzed the performance of the model with the fine-tuning and the performance of it without the fine-tuning. This way, we can easily see the effect of the fine-tuning.

The following I copied from the answer of Jonathan Uesato on Quora, it explains very well:

An ablation study typically refers to removing some “feature” of the model or algorithm and seeing how that affects performance.
  • An LSTM has 4 gates: feature, input, output, forget. We might ask: are all 4 necessary? What if I remove one? Indeed, lots of experimentation has gone into LSTM variants, the GRU being a notable example (which is simpler).
  • If certain tricks are used to get an algorithm to work, it’s useful to know whether the algorithm is robust to removing these tricks. For example, DeepMind’s original DQN paper reports using (1) only periodically updating the reference network and (2) using a replay buffer rather than updating online. It’s very useful for the research community to know that both these tricks are necessary, in order to build on top of these results.
  • If an algorithm is a modification of a previous work, and has multiple differences, researchers want to know what the key difference is.
  • Simpler is better (inductive prior towards simpler model classes). If you can get the same performance with two models, prefer the simpler one.


  1. Replies
    1. Five weeks ago my boyfriend broke up with me. It all started when i went to summer camp i was trying to contact him but it was not going through. So when I came back from camp I saw him with a young lady kissing in his bed room, I was frustrated and it gave me a sleepless night. I thought he will come back to apologies but he didn't come for almost three week i was really hurt but i thank Dr.Azuka for all he did i met Dr.Azuka during my search at the internet i decided to contact him on his email he brought my boyfriend back to me just within 48 hours i am really happy. What’s app contact : +44 7520 636249‬

  2. Really Good blog post. provided a helpful information. I hope that you will post more updates like this Data Science online Training Bangalore

  3. I would like to thank you for the efforts you have made in writing this article. I am hoping the same best work from you in the future as well. In fact your creative writing abilities has inspired me to start my own BlogEngine blog now. Really the blogging is spreading its wings rapidly. Your write up is a fine example of it.Jogos online
    friv free Games
    play Games friv 2020


    It is a very hard situation when playing the lottery and never won, or keep winning low fund not up to 100 bucks, i have been a victim of such a tough life, the biggest fund i have ever won was 100 bucks, and i have been playing lottery for almost 12 years now, things suddenly change the moment i came across a secret online, a testimony of a spell caster called DR EMU, who help people in any type of lottery numbers, i was not easily convinced, but i decided to give try, now i am a proud lottery winner with the help of DR EMU, i won $1,000.0000.00 and i am making this known to every one out there who have been trying all day to win the lottery, believe me this is the only way to win the lottery.

    Contact him via email
    What's app +2347012841542

  5. This post is very simple to read and appreciate without leaving any details out. Great work! omnichannel contact center solution

  6. Ablation study in machine learning refers to a systematic analysis that aims to understand the contribution or impact of individual components or features within a machine learning model. It involves selectively removing or disabling specific components of the model to observe how it affects the overall performance. By conducting an ablation study, cheap camera for photography, researchers can gain insights into the importance and effectiveness of different parts of the model and make informed decisions regarding its design and optimization.

    In the context of a commenting website, an ablation study could be conducted to evaluate the significance of various factors influencing the comment classification or sentiment analysis model. For example, researchers might analyze the effect of different features such as the length of the comment, the presence of specific keywords, or the inclusion of contextual information. By selectively removing these features and observing the impact on the model's performance, they can assess which features contribute the most to accurate comment analysis.

    Ablation studies can also be used to investigate the performance of different model architectures or algorithms. For instance, in the case of a machine learning model used for comment recommendation, best mirrorless camera, researchers might evaluate the effectiveness of different recommendation algorithms by disabling or substituting them one by one. This analysis helps identify the most influential algorithm or component and guides the selection of the most appropriate recommendation strategy for the commenting website.

  7. itu bagus, terimakasih informasinya.. jika anda mencari informasi tentang kursi roda silahkan cek di
    Kursi Roda Traveling dan
    Harga Kursi Roda Terbaru