The hard thing about deep learning

The hard thing about deep learning  #ai

  • In a nutshell: the deeper the network becomes, the harder the optimization problem becomes.
  • The simplest neural network is the single-node perceptron , whose optimization problem is convex .
  • To provably solve optimization problems for general neural networks with two or more layers, the algorithms that would be necessary hit some of the biggest open problems in computer science.
  • There is a rich variety of optimization algorithms to handle convex optimization problems, and every few years a better polynomial-time algorithm for convex optimization is discovered.
  • Judd also shows that the problem remains NP-hard even if it only requires a network to produce the correct output for just two-thirds of the training examples, which implies that even approximately training a neural network is intrinsically difficult in the worst case.

Deeper neural nets often yield harder optimization problems.
Continue reading “The hard thing about deep learning”

Using Machine Learning to Detect Malicious URLs

#ICYMI Using #MachineLearning to Detect Malicious URLs  #security

  • Now that we have the data in our list, we have to vectorize our URLs.
  • KDnuggets Home > News > 2016 > Oct > Opinions, Interviews > Using Machine Learning to Detect Malicious URLs ( 16:n39 )
  • I wrote my own tokenizer function for this since URLs are not like some other document text.
  • Unfortunately or fortunately, there has been little work done on security with machine learning algorithms.
  • The data and code is available at Github .


This is a write-up of an experiment employing a machine learning model to identify malicious URLs. The author provides a link to the code for you to try yourself.

Continue reading “Using Machine Learning to Detect Malicious URLs”

Mapping global fishing activity with machine learning

  • Global Fishing Watch combines cloud computing technology with satellite data to provide the world’s first global view of commercial fishing activities.
  • Today, Global Fishing Watch is an early preview of what is possible.
  • Indonesia’s Minister of Fisheries and Marine Affairs, Susi Pudjiastuti, has committed to making the government’s Vessel Monitoring System (VMS) public in Global Fishing Watch in 2017.
  • Global Fishing Watch was not possible five years ago.
  • Bali Seafood, the largest exporter of snapper from Indonesia, has teamed up with Pelagic Data Systems, manufacturers of cellular and solar powered tracking devices to bring the same transparency for small scale and artisanal fishing vessels, into Global Fishing Watch as part of a pilot program.

Global Fishing Watch is a free, simple, online platform that gives anyone, anywhere, a way to to visualize, track, and share information about fishing activity worldwide.
Continue reading “Mapping global fishing activity with machine learning”

A Development Methodology for Deep Learning – Medium

A Development Methodology for #DeepLearning.  #BigData #MachineLearning #DataScience #AI

  • The machine learning convention has been to create a training set, a validation set and a test set.
  • Although Deep Learning is built from software it is a different kind of software and a different kind of methodology is needed.
  • The observations that differs from conventional machine learning is that Deep Learning has more flexibility in that a developer has the additional options of employing either a bigger model or using more data.
  • The methodology addresses the necessary interplay of the need for more training data and the exploration of alternative Deep Learning patterns that drive the discovery of an effective architecture.
  • Deep Learning differs most from traditional software development in that a substantial portion of the process involves the machine learning how to achieve objectives.

The practice of software development has created development methodologies such agile development and lean methodology to tackle the…
Continue reading “A Development Methodology for Deep Learning – Medium”

Using Machine Learning to Name Malware

Using #MachineLearning To Name #Malware. #BigData #DataScience #AI

  • All of the post-processing to settle on one common name has already been done and you can find the library that can guess the virus names at this github repo .
  • Using Machine Learning to Name Malware – Artificial Intelligence on Using Machine Learning to Name Malware
  • items() if v} print(“We have to guess the family name in the following result:\n”) print(to_guess) l_of_l = get_list_of_token_lists([to_guess]) m = tfidf.transform(l_of_l) els_to_pos = {e: tfidf.vocabulary_[e] for e in l_of_l[0]} els_to_scores = {k: m[:, v].
  • “We have to guess the family name in the following result:” {‘AVG’: ‘MLoader’, ‘Ad-Aware’: ‘Gen:Application.
  • If we just see the eggs and we know the probabilities ahead of time, we can figure out which egg belongs to which dinosaur using Viterbi algorithm .

Extracting information from malware names using Conditional Random Fields.
Continue reading “Using Machine Learning to Name Malware”

Artificial Intelligence: Don’t Fear It, Embrace It

Artificial Intelligence: Don't Fear It, Embrace It | #BigData #Artificialintelligence #RT

  • Many intermediary steps had to be taken to teach machine learning systems.
  • “We’re seeing the point at which data-driven deep learning systems are starting to overtake systems that we’ve engineered ourselves,” said Coates. “
  • Big data could drive the next big security strategy shift.
  • Thanks to mobile devices, data that’s useful to deep learning systems is being generated at an ever growing rate.
  • But traditional machine learning hit a wall, Coates said.

Adam Coates, the director of the Baidu Research’s Silicon Valley AI Lab, says don’t fear artificial intelligence. Instead, look to it to save lives. He spoke at the InformationWeek Elite 100 Conference this week.
Continue reading “Artificial Intelligence: Don’t Fear It, Embrace It”

Using Machine Learning to Name Malware

Using #MachineLearning To Name #Malware. #BigData #DataScience #AI #Cybersecurity

  • All of the post-processing to settle on one common name has already been done and you can find the library that can guess the virus names at this github repo .
  • items() if v} print(“We have to guess the family name in the following result:\n”) print(to_guess) l_of_l = get_list_of_token_lists([to_guess]) m = tfidf.transform(l_of_l) els_to_pos = {e: tfidf.vocabulary_[e] for e in l_of_l[0]} els_to_scores = {k: m[:, v].
  • “We have to guess the family name in the following result:” {‘AVG’: ‘MLoader’, ‘Ad-Aware’: ‘Gen:Application.
  • Using Machine Learning to Name Malware
  • Let’s create some training data to label parts of virus names with their corresponding tags.

Extracting information from malware names using Conditional Random Fields.
Continue reading “Using Machine Learning to Name Malware”

Artificial Intelligence: Don’t Fear It, Embrace It

Artificial Intelligence: Don't Fear It, Embrace It | #BigData #Artificialintelligence #RT

  • Many intermediary steps had to be taken to teach machine learning systems.
  • “We’re seeing the point at which data-driven deep learning systems are starting to overtake systems that we’ve engineered ourselves,” said Coates. “
  • Big data could drive the next big security strategy shift.
  • Thanks to mobile devices, data that’s useful to deep learning systems is being generated at an ever growing rate.
  • But traditional machine learning hit a wall, Coates said.

Read the full article, click here.


@Ronald_vanLoon: “Artificial Intelligence: Don’t Fear It, Embrace It | #BigData #Artificialintelligence #RT”


Adam Coates, the director of the Baidu Research’s Silicon Valley AI Lab, says don’t fear artificial intelligence. Instead, look to it to save lives. He spoke at the InformationWeek Elite 100 Conference this week.


Artificial Intelligence: Don’t Fear It, Embrace It