News for hackers. Stuff that matters.

iamaaditya

iamaaditya on Hacker News

Re:Waymo halts service during S.F. blackout after causing traffic jams on Sunday December 21, 2025 @05:58PM (#46349502)
One usage case that I saw myself is when a vehicle is parked such that it will require the other vehicle to go slighty over the curb, in this case the curb is flat so I assuming the parked driver thought it was okay. Every other human driver did okay, but Waymo just refused to put its wheel on the curb and just got stuck. Video here: https://x.com/aaditya_prakash/status/1989444130238259575?s=2...
Re:Accelerated PyTorch Training on M1 Mac on Wednesday May 18, 2022 @02:14PM (#31425887)
In machine learning (especially deep learning or neural networks), the 'training' is done by using Stochastic Gradient Descent. These gradients are computed using Backpropagation. Backpropagation requires you to do a backward pass of your model (typically many layers of neural weights) and thus requires you to keep in memory a lot of intermediate values (called activations). However, if you are doing "inference" that is if the goal is only to get the result but not improve the model, then you don't have to do the backpropagation and thus you don't need to store/save the intermediate values. As the layers and number of parameters in Deep Learning grows, this difference in computation in training vs inference becomes signifiant. In most modern applications of ML, you train once but infer many times, and thus it makes sense to have specialized hardware that is optimized for "inference" at the cost of its inability to do "training".
Re:Show HN: Wortharead – Find intelligent blogs and newsletters on Thursday February 25, 2021 @03:23PM (#26267768)
Love the UI and the simple design. All the best with the product!
Re:A Nepali team has made the first winter ascent of K2 on Saturday January 16, 2021 @04:57PM (#25805886)
I believe there is some form of memory bias or bias in reporting by the media. If you look at the record books, you will find that most records are held by Sherpas. For example, more than 70 records for Mount Everest is held my Sherpas (or other Nepalese climbers) [1], far more than climbers from any other country.

Apa Sherpa, one of the most prolific Sherpas, holds many records [2]. Interesting fact he was Sherpa to Peter Hillary (Edmund's son).

[1] https://en.wikipedia.org/wiki/List_of_Mount_Everest_records [2] https://en.wikipedia.org/wiki/Apa_Sherpa

Re:Ron Graham has died on Tuesday July 07, 2020 @10:08PM (#23766263)
Graham's Number, as in 3 ↑↑ 3, is named after Ron Graham.

[1] https://plus.maths.org/content/too-big-write-not-too-big-gra...

[2] https://en.wikipedia.org/wiki/Graham%27s_number?oldformat=tr...

[3] And here is the man himself describing it https://www.youtube.com/watch?v=GuigptwlVHo

Re:Learning Math for Machine Learning on Thursday August 02, 2018 @06:00PM (#17675838)
[The Matrix Calculus You Need For Deep Learning](https://arxiv.org/pdf/1802.01528.pdf)
Re:Attacks against machine learning – an overview on Sunday June 10, 2018 @05:57PM (#17280390)
One simple way to minimize impact of these attacks is our work called Pixel Deflection (CVPR 2018 Spotlight). Here is a short (4 min) video introduction to the idea https://youtu.be/VgjOXJ9QKWo
Re:Pixel Deflection – a simple transformation based defense on Thursday February 01, 2018 @08:45AM (#16281771)
Paper: https://arxiv.org/abs/1801.08926

Code: https://github.com/iamaaditya/pixel-deflection

Re:10GHz at under 1V by 2005 - The future of Intel’s manufacturing pro... on Wednesday February 08, 2017 @12:22AM (#13595989)
Technically it was heat but mostly it was due to (i) Economics, there was less demand for faster clock speed. Otherwise more research could have gone towards solving heat problem. (ii) Each cycle of CPU was more efficient with ability to execute multiple instructions in a single cycle and with more efficient instruction sets.

Surprisingly, power consumption also made huge impact. As tablets and laptops got more popular than desktop battery life became a major concern and thus TDP played major role in research.

Try this fun experiment: Underclock your CPU by half a GHz and see if you notice the difference in your day to day work.

Re:H-1B Visas: U.S. Lawmaker Re-Introduces Bill to Tighten Rules on Monday January 09, 2017 @08:28PM (#13361725)
Recently, I have seen O1 being given pretty generously. Couple of people who I know who completed PhDs from mediocre University and less than 10 papers in total have all gotten their O1s. Yes they had to get reference letters and complete all the formality but nothing stellar or extra-ordinary was required.
Re:Faster Images using HTTP2 and Progressive JPEGs on Wednesday December 28, 2016 @10:55AM (#13271260)
If it would be sufficient to have high PSNR only in the 'regions of interest', then you could use something like this [1], which uses a CNN model to predict a map, and a multiple JPEG encoding to achieve variable quality compression.

[1] https://github.com/iamaaditya/image-compression-cnn

Re:Image-to-Image Translation with Conditional Adversarial Nets on Tuesday November 22, 2016 @02:08PM (#13016590)
Here is my work, where I do use semantic information to achieve compression (rather improve JPEG). This is not an end to end compression like Google's work, but just incorporating semantic knowledge into compression. I am still trying to clean up the code before I make arxiv/github submission, but since you are interested here is the link http://gpgpu.cs-i.brandeis.edu/semantic_jpeg.pdf
Re:Optimized-Torch – Torch installation in a self-contained folder on Monday September 05, 2016 @09:44PM (#12433192)
This library does not bundle Intel MKL. It is a prerequisite.
Re:“My wife has complained that OpenOffice will never print on Tuesday... on Tuesday May 17, 2016 @07:25PM (#11718464)
https://support.microsoft.com/en-us/kb/3053711

(from the link)

Resolution

To resolve the issue, do not create a user account contains the string "user" on the computer.

Re:End to End Learning for Self-Driving Cars [pdf] on Tuesday May 10, 2016 @01:05PM (#11668718)
This reminds me of ensemble. Most machine learning techniques benefit in accuracy and precision by ensembling different methods. More different the methods are (variance) better improvement you get.
Re:Deep learning for visual question answering: demo with Keras code on Wednesday April 06, 2016 @12:35AM (#11436642)
hi syllogism

Right now everyone doing this is highly focussed on the competition and trying to beat the numbers. For that purpose certainly they would want to stick to predicting Top K answers.

For e.g see this table

  Model 	Q+I [1]	 Q+I+C [1] 	ATT 1000 	ATT Full

  ACC. 	0.2678 	0.2939 		0.4838 		0.4651
Where ATT full is when using all the words, it performs worse than ATT 1000 (Source)[Chen, K., Wang, J., Chen, L. C., Gao, H., Xu, W., & Nevatia, R. (2015). ABC-CNN: An Attention Based Convolutional Neural Network for Visual Question Answering. arXiv preprint arXiv:1511.05960.]

Once the competition is over (in rough two months), there will be more focus on actual AI part, where generating the answers would be the right thing to do. There are other papers where they use external knowledge base like DBPedia, certainly "answer word" could be picked up from there.

What you have suggested is a very interesting approach, and I am not aware of any paper which has tried that. Certainly quite a few paper have tried to extend NLP QA to Visual QA but with limited success (expect Metamind people). I will certainly keep that in my ideas to try list. I will update you if I get some results.

P.S: Thank you for creating Spacy, I love it and I use it everyday !

Re:Deep learning for visual question answering: demo with Keras code on Wednesday April 06, 2016 @12:23AM (#11436587)
1.

   Model 	Q+I [1]	 Q+I+C [1] 	ATT 1000 	ATT Full 
   ACC. 	0.2678 	0.2939 		0.4838 		0.4651
Where ATT Full represents using all the words in the vocabulary, as you can see it performs worse than "Most frequent 1000 answers".

Source:

   Chen, K., Wang, J., Chen, L. C., Gao, H., Xu, W., & Nevatia, R. (2015). ABC- CNN: An Attention Based Convolutional Neural Network for Visual Question Answering. arXiv preprint arXiv:1511.05960.
2.

(a) Several early papers about VQA directly adapt the image captioning models to solve the VQA problem [10][11] by generating the answer using a recurrent LSTM network conditioned on the CNN output. But these models’ performance is still limited [10][11]

(b) our own implementation of this model is less accurate on [2] than other baseline models

Above two quotes are from -

   Xu, Huijuan, and Kate Saenko. "Ask, Attend and Answer: Exploring Question-Guided Spatial Attention for Visual Question Answering." arXiv preprint arXiv:1511.05234(2015).
However, I think my words were sloppy, as I could not find more concrete proof in the literature, but I will revisit them with detail to recollect where I read about RNN generating answers not overachieveing softmax classification over Top K distribution of answers.

Also, I would like to note that, I am not using only "one word answers" as the possible set of answers. It contains few two words, and very few three and four word answers.

Here is the distribution Key == Length of words | Value == Count of answers with those many words

Counter({1: 855, 2: 112, 3: 32, 4: 1})

Re:Deep learning for visual question answering: demo with Keras code on Wednesday April 06, 2016 @12:16AM (#11436561)
So, I experimented with Skip-thought vectors (as question), to embody the the semantic knowledge about the question but that performed poorly, I mean very poorly. Tbh, I was surprised with that.

Currently the way system works, is highly biased towards the individual words in the question, whenever it sees the word "color" in the question it is going to pick one of the 'colors' as the answer, even if the question is not asking about colors. This is certainly due all the priors from training data.

Re:Deep learning for visual question answering: demo with Keras code on Tuesday April 05, 2016 @05:41PM (#11434757)
Even though this is "Question Answering", it is trained as a classification model. Thus the model will try to come up with one of the top "1000" answers it has seen during the training. This certainly limits the possibility of answers and sometimes returns very weird answer.

It is not for lack of trying that all the top papers in visual question answering end up doing this as a classification task. Results are really poor when it is used as RNN generation, and also extending it more than top 1000 answers does not yield any better results. 87% of the questions in training + validation is within 1000 unique answers.

Latest models have started using more complex form of memory and more tightly integrating the question vectors. One of the top model called DPPNet trains a separate matrix from the question vector (chain of GRUs) to find correspondence on the image filter weights. Their idea is that some question have more relevant areas in the image features. Yet another model DMN+, by Metamind uses dynamic memory network which they build to do language question answers but the extension to images work pretty good.

Surprisingly the models that use visual attention are not the best and I think it is mostly because this kind of model requires even more data and longer training. Just taking 10 different crop of the question image and doing voting of answer beats attention models (based on numbers reported by these papers).

Right now I am working on converting "End to end network" -http://arxiv.org/abs/1503.08895 to this task. I tried working on Neural Turing machine but I could not make it work for this kind of task, but it was mostly because of lack of indepth understanding of NTM.

Any feedback from you guys are welcome.

P.S Thanks fchollet for writing Keras and for this post. Can't wait to try Keras 1.0

Re:Deep learning for visual question answering: demo with Keras code on Tuesday April 05, 2016 @05:23PM (#11434613)
Hi Harperlee

I ran the question you asked --

"Are they playing tennis?"

and following is the result

99.93 % yes

00.05 % no

000.0 % right

000.0 % left

000.0 % black and white