Scientists are getting closer to building life from scratch and technology pioneers are taking notice, with record sums moving into a field that could deliver novel drugs, materials, chemicals and even perfumes. Despite ethical and safety concerns, investors are attracted by synthetic biology's wide market potential and the plummeting cost of DNA synthesis, which is industrialising the writing of the genetic code that determines how organisms function. While existing biotechnology is already used to make medicines like insulin and genetically modified crops, synthesising whole genes or genomes gives an opportunity for far more extensive changes. Alexander the friendly robot visits the Indoor Park in London (file pic). Technology pioneers are investing huge sums of money in all manner of innovative ideas.

"The intersection of biology and technology is a difficult place to be because of different cultures and languages, but I think we are breaking through some of those barriers," said Thomas Bostick, former head of the U.S. Army Corps of Engineers who now leads biotech firm Intrexon's environment unit.

Of the many great things promised by Crispr gene editing technology, the ability to eliminate disease by modifying organisms might just top the list. But doing that requires perfecting something called a gene drive. Think of gene drives as a means of supercharging evolution to, say, give an entire population of mosquitoes a gene that kills the Zika virus. The trouble is, organisms develop resistance to gene drives, much like they eventually outwit pesticides and antibiotics. Researchers dedicate no small amount of time and thought to creating gene drives that can outsmart evolution because the potential payoffs are so great.

Traditional Linear Genetic Programming (LGP) algorithms are based only on the selection mechanism to guide the search. Genetic operators combine or mutate random portions of the individuals, without knowing if the result will lead to a fitter individual. Probabilistic Model Building Genetic Programming (PMB-GP) methods were proposed to overcome this issue through a probability model that captures the structure of the fit individuals and use it to sample new individuals. This work proposes the use of LGP with a Stochastic Context-Free Grammar (SCFG), that has a probability distribution that is updated according to selected individuals. We proposed a method for adapting the grammar into the linear representation of LGP. Tests performed with the proposed probabilistic method, and with two hybrid approaches, on several symbolic regression benchmark problems show that the results are statistically better than the obtained by the traditional LGP.

I watched Ilya Sutskever's talk on their new evolutionary strategies paper. The reason this paper is fascinating is that they use a relatively dumb, simple stochastic method of optimisation that shouldn't really work well in practice, and show that it is actually competitive with SGD/back-propagation-based methods in RL. This is mainly due to the fact that it parallelizes so naturally. Evolution strategies (ES) is can be best described as a gradient descent method which uses gradients estimated from stochastic perturbations around the current parameter value. While the authors did comparisons in the context of RL, and there are many RL-specific advantages, here I'm focussing on ES as a general black-box optimisation method.

It's not easy to train a neural network. Even if they're not difficult to implement, it can take hours to get them ready no matter how much computing power you can use. OpenAI researchers may have a better solution: forgetting many of the usual rules. They've developed an evolution strategy (no, it doesn't relate much to biological evolution) that promises more powerful AI systems. Rather than use standard reinforcement training, they create a "black box" where they forget that the environment and neural networks are even involved.

We present mlrMBO, a flexible and comprehensive R toolbox for model-based optimization (MBO), also known as Bayesian optimization, which addresses the problem of expensive black-box optimization by approximating the given objective function through a surrogate regression model. It is designed for both single- and multi-objective optimization with mixed continuous, categorical and conditional parameters. Additional features include multi-point batch proposal, parallelization, visualization, logging and error-handling. mlrMBO is implemented in a modular fashion, such that single components can be easily replaced or adapted by the user for specific use cases, e.g., any regression learner from the mlr toolbox for machine learning can be used, and infill criteria and infill optimizers are easily exchangeable. We empirically demonstrate that mlrMBO provides state-of-the-art performance by comparing it on different benchmark scenarios against a wide range of other optimizers, including DiceOptim, rBayesianOptimization, SPOT, SMAC, Spearmint, and Hyperopt.

I read Artificial Intelligence for Humans, Volume 1 and then ordered volumes 2 and 3. What I like about this series is the same thing I like about Volume 2, that it's very readable. For someone without a math background, and limited programming prowess, I can understand the concepts. The only things about the book that I don't like are: 1) Some of the context is missing. For instance, I can understand Genetic Algorithms, Partical Swarm Optimization, and Ant Colony Optimization as concepts and I think I could basically code them if I needed to. I would say his forte is explaining the ideas and the math in plain language.

The emergence and development of cancer is a consequence of the accumulation over time of genomic mutations involving a specific set of genes, which provides the cancer clones with a functional selective advantage. In this work, we model the order of accumulation of such mutations during the progression, which eventually leads to the disease, by means of probabilistic graphic models, i.e., Bayesian Networks (BNs). We investigate how to perform the task of learning the structure of such BNs, according to experimental evidence, adopting a global optimization meta-heuristics. In particular, in this work we rely on Genetic Algorithms, and to strongly reduce the execution time of the inference -- which can also involve multiple repetitions to collect statistically significant assessments of the data -- we distribute the calculations using both multi-threading and a multi-node architecture. The results show that our approach is characterized by good accuracy and specificity; we also demonstrate its feasibility, thanks to a 84x reduction of the overall execution time with respect to a traditional sequential implementation.

Machine Learning (ML), a branch of Computer Science that focuses on drawing insights and conclusions by examining data sets, is an increasingly popular discipline today in resolving enterprise business issues. However the field is vast and consists of numerous algorithms and approaches. Data sets are also often complex and require to be pre-processed before an ML algorithm can be'trained' to learn from such data. For a particular problem domain and data set, defining the pre-processing technique and selecting the ML algorithm (or set of algorithms) is still largely'an art rather than a science' depending on the knowledge and skills of the expert/data scientist in question. With time this will change and scientific guiding principles/best practices will emerge to pre-process data and to select appropriate algorithms for a particular problem domain - as the discipline matures.