The increasing size of large language models (LLMs) challenges their usage on resource-constrained platforms. For example, memory on modern GPUs is insufficient to hold LLMs that are hundreds of Gigabytes in size. Offloading is a popular method to escape this constraint by storing weights of an LLM model to host CPU memory and SSD, then loading each weight to GPU before every use. In our case study of offloaded inference, we found that due to the low bandwidth between storage devices and GPU, the latency of transferring large model weights from its offloaded location to GPU memory becomes the critical bottleneck with actual compute taking nearly 0% of runtime. To effectively reduce the weight transfer latency, we propose a novel sparse format that compresses the unstructured sparse pattern of pruned LLM weights to non-zero values with high compression ratio and low decompression overhead. Endor achieves this by expressing the positions of non-zero elements with a bitmap. Compared to offloaded inference using the popular Huggingface Accelerate, applying Endor accelerates OPT-66B by 1.70x and Llama2-70B by 1.78x. When direct weight transfer from SSD to GPU is leveraged, Endor achieves 2.25x speedup on OPT-66B and 2.37x speedup on Llama2-70B.

Background: Tree ensemble machine learning models are increasingly used in microbiome science as they are compatible with the compositional, high-dimensional, and sparse structure of sequence-based microbiome data. While such models are often good at predicting phenotypes based on microbiome data, they only yield limited insights into how microbial taxa or genomic content may be associated. Results: We developed endoR, a method to interpret a fitted tree ensemble model. Both the network and importance scores derived from endoR provide insights into how features, and interactions between them, contribute to the predictive performance of the fitted model. Adjustable regularization and bootstrapping help reduce the complexity and ensure that only essential parts of the model are retained. We assessed the performance of endoR on both simulated and real metagenomic data.

Since the launch in 2008, the phenomenon, cryptocurrency, has taken the financial world by storm. The market that started with Bitcoin has now more than 3000 cryptocurrencies. As of June 2021, the total market cap of cryptocurrencies is US$1,746,285,217,570, but this success came as a result of a lot of volatility. Bitcoin, alone, fluctuated from its peak of US$60,000 to around US$30,000 recently. Over time, investors observed this volatility and realized that a lot of money can be made through crypto investments.

One can argue that even the most innovative banking institutions are bureaucratic enough, and their slow decision-making causing banks to lose their premium over fintech applications, peer to peer lending marketplaces, and payment processors. At the same time, many expanded into the business of micro-lending. Banking services are no longer a monopoly of banks, and traditional financial institutions have to innovate in order to survive. The era of non-traditional financial services providers such as Amazon Payments, PayPal Payments and PayU, has risen. The launch of the Payment Services Directive II in Europe unlocks new dynamics for FinTech and Payment Services.

Big data is currently on everybody's lips with stringent regulations the order of the day and security breaches happening on a regular basis. A company called Endor has come up with a blockchain and AI based solution to manage and process data. After years of research at MIT, Endor claims to have invented the "Google for predictive analytics*", providing automated AI predictions for companies. Endor can process Encrypted Data, without ever decrypting it, on and off blockchain and it enables business users to ask predictive questions and get automated accurate predictions. No data science expertise is required.

Big data is currently on everybody's lips with stringent regulations the order of the day and security breaches happening on a regular basis. A company called Endor has come up with a blockchain and AI based solution to manage and process data. After years of research at MIT, Endor claims to have invented the "Google for predictive analytics*", providing automated AI predictions for companies. Endor can process Encrypted Data, without ever decrypting it, on and off blockchain and it enables business users to ask predictive questions and get automated accurate predictions. No data science expertise is required.

Big data analytics and machine learning are on the rise and set for massive further growth over the coming years. The results of a survey conducted jointly by MIT Technology Review and Google Cloud showed that 60 percent of respondents have already implemented a machine learning strategy in their organization. Furthermore, Deloitte predicts that spending on machine learning (ML) and AI will nearly quadruple from $12 billion in 2017 to $57.6 billion in 2021. Amidst this growing popularity, a growing concern is that algorithms are only as good as the data that's fed into them. The old adage "garbage in, garbage out" applies to AI and ML as much as it does to any other computing-based system.

To supplement and even compete with the intelligence of human beings, many technological developments have come into existence to enhance Artificial Intelligence over the years. After a long experimental struggle, developers within the space, have formed systems whose artificial intelligence is demonstrating increasing promise. One such development using artificial intelligence is blockchain. Read on to find out why this collaboration of technologies could have a transformative outcome for all. Blockchain is the technology that stores encrypted data known as blocks in electronic ledgers.

Artificial intelligence, or AI for short, has become a real buzzword these last couple of years. No longer restricted to the fanciful realm of science fiction, artificial intelligence is being improved upon each day in very real ways. And this includes its increasingly advanced and widespread application in the context of content creation. Here are a few examples of companies and industry experts that are exercising the power of artificial intelligence in creative ways to generate new and engaging online content. It's fair to say that Google is top dog when it comes to modern search engines.

Hundreds of new cryptocurrencies have been created and offered to investors through initial coin offerings (ICOs) over the past year. Millions of new users entered the crypto space in 2017 during this ICO boom. More are jumping on the bandwagon this year. Most people who've heard of cryptocurrencies – and many who have put money into it – only have a vague understanding of how these work as investment vehicles. Confusion among new investors has been high due to the abundance of coins and their fluctuating valuations.