naresh_xai

6 ms·

1.
by naresh_xai 5y ago
I think it can be done. Just maintain the data not in a standard tabular format but a time based graph instead. And put constraints to not look back. Constrained deep learning is a thing too.
2.
by naresh_xai 5y ago
Current set of tools in explainable AI do allow deep neural nets to tell why they gave the response that they did to a certain extent. Contact me if you want to see it applied to text based DL models, time series based DL models or Image ba
3.
by naresh_xai 5y ago
Current set of tools in explainable AI do allow deep neural nets to tell why they gave the response that they did to a certain extent. Contact me if you want to see it applied to text based DL models, time series based DL models or Image ba
4.
by naresh_xai 5y ago
Are we still limiting to visual cues and not the auditory,smell,taste,touch data which we get exposed to?
5.
by naresh_xai 5y ago
Umm, no there are clear verification methods for Explainable AI techniques today. One way to check the justification would be if things which were important in the justification were removed in some sense, then would the output change signf
6.
by naresh_xai 5y ago
That’s basically the gist of explainable AI
7.
by naresh_xai 5y ago
Pretty cool stuff
8.
by naresh_xai 5y ago
If revenue was more in a non-Apple ecosystem, would it not be the case that devs would all move to a non-Apple ecosystem. Android also has the problem that users are far less likely to be as educated (since target audience is not white coll
9.
by naresh_xai 5y ago
But apple does not allow for that. So in that sense, they do not allow businesses for targeted advertisement. Please correct me if I am wrong. Their processes and org structure is not built for that either with siloed teams. So much that qu
10.
by naresh_xai 5y ago
And 2 weeks later, the guy will resell stuff from another account from different paid reviews. Which are also always evolving to capture consumer trust. Works especially well for products in usd 1k range.
11.
by naresh_xai 5y ago
Companies list/re-list products on amazon faster than you can say ‘go’ unfortunately.
12.
by naresh_xai 6y ago
The number of loans a bank gives out to such a specific factor might be low enough that it would take 30 years for a bank to have an actual study. That's 1-2 generations of abuse. Most banks just copy each other.
13.
by naresh_xai 6y ago
Unless you have causal proof, its irresponsible for a business to use such factors in modeling outcomes.
14.
by naresh_xai 6y ago
I also think that it is possible that the model learned that information from too small of a data sample. What is a good data sample for every such feature in a relatively balanced manner is really difficult to build a dataset from. Conside
15.
by naresh_xai 6y ago
AI with no dibiasing does not work often enough. I’ve seen enough examples in Computer Vision models to say so. (Even segmentation models tend to rely on external clues and this affects the generalizability of the model). Happy to share mor
16.
by naresh_xai 6y ago
Happy to share a lot more ethical and challenges of non transparent AI via email with you. Bigger companies are pretty much hiding behind non transparent models and trying to ignore their failures at the moment.
17.
by naresh_xai 7y ago
Exactly why we need causal reasoning/causal proof alongwith ML
18.
by naresh_xai 7y ago
Interested in seeing where symbolic AI crowd has disagreed with that. Only group of people who disagree that I know of are a small set of people who think you should build inherently explainable models as opposed to explaining the decisions
19.
by naresh_xai 7y ago
Happy to share decks and references to your email address. Please share your email address with me :)
20.
by naresh_xai 7y ago
Graph Neural networks are currently used a lot in the neural networks for drug discovery space. They significantly beat RNN and CNN baseline/complex versions equivalents on the same datasets(Tox21, QM9 efc).
21.
by naresh_xai 7y ago
@Paul Robinson: Did you look at the number of mislabeled images in the udacity self driving car dataset? If you do not have an understanding of the subject yourself, you’re likely to make the same annotation errors which feed into a bad mod
22.
by naresh_xai 7y ago
Think more in terms of function decomposition rather than having to look at each individual parameter and you will find papers and techniques which lead to deep neural networks being quite explainable. Contrastive LRP would be a good starti
23.
by naresh_xai 7y ago
Again the non-sequitur argument of an explainable model must be worse than a deep learning system and there has to be a tradeoff. You don’t need to reduce complexity to induce explainability. You just need to decompose the function into sma
24.
by naresh_xai 7y ago
You should definitely look at advances in the field then. There a lot more promising work beyond grad cam. And there are a few techniques for human oriented explanations - namely TCAV (human oriented explanations) and PatternNet/Patter
25.
by naresh_xai 7y ago
Not to mention the fact that most models are trained without a background class and tend to give overconfident predictions on out of distribution samples.
26.
by naresh_xai 7y ago
Let’s see. Robotics problems were claimed to be tractable through AI. However, a large majority of robotics solutions today are 90% derived through control systems (which follow some degree of causal analysis) followed by AI to optimize the
27.
by naresh_xai 7y ago
Or you can actually visualize and quantify impact of aggregates of meaningful features within networks using the methodology described in NetDissect or in TCAV. But of course, that casts doubt into a lot of claimed mechanisms and tons of ML
28.
by naresh_xai 7y ago
If you were ever involved in the drug discovery process, then you would know that statistical evidence through clinical trials is only for the last ounce of drug testing. Out of 10,000+ drug molecule candidates only ~ 5 molecules get select
29.
by naresh_xai 7y ago
Which is strongly misleading tradeoff. For a ton of tasks, deep learning methods are no better than white box regression or tree ensemble methods. And there is no reason to expect that a deep learning model has to be unexplainable. He’s put
30.
by naresh_xai 7y ago
When a person’s claim to fame and research is dependent on ignoring explainability and causality in research, he will ignore it to the best of his means. To him, all the precursors to clinical trials (selection of a molecule from restricted

More ›