Caravaggio 2025 at Rome’s Palazzo Barberini

According to the Caravaggio 2025 exhibition literature, it was 74 years to the day (April 20, 1951) from the initial Caravaggio exhibition to the beginning of the one currently being held at the Palazzo Barberini in Rome.

Palazzo Barberini

The current exhibition was slated to conclude on July 7 but, due to its great success, has been extended for an additional two weeks. I had resisted the temptation to make the trip but once it was extended, I could resist no longer. I visited the exhibition on July 18.

The exhibition included 24 (of approximately 73 undisputed) Caravaggio’s paintings drawn from all corners of the earth. The extended exhibition did not include the Metropolitan Museum’s Concert as it had been previously committed for that period.

The exhibition was spread over four rooms in the museum and was designed to trace the painter’s evolution over the 15-year period covering his arrival in Rome in 1595 and his death in Porto Ercole in 1610.

Each of the works are noteworthy but a few merited especial attention:

• Portrait of Maffeo Barberini (1598 - 1599) — This painting was published by Longhi in 1963 but has never been displayed to the public
• Ecce Homo — this painting was recently rediscovered and, with this exhibition, has returned to Italy after many centuries (patrons were not allowed to take pictures of this painting)
• The first version of the Conversion of Saul for the Cerasi chapel. This painting is not readily accessible to the public as it is privately owned.

Caravaggio, Portrait of Maffeo Barberini,
1598 - 1599 c.

Caravaggio, Conversion of Saul, 1600 - 1601

It was a pleasure to see these works; especially after having seen three of Caravaggio's paintings at the National Gallery 6 months earlier. None of the National Gallery paintings that I had seen were included in the exhibition.

As impressive as the lineup was, there were a few display issues which, in my opinion. somewhat tainted the show. First, the lighting was not up to the standards of a museum exhibiting a blockbuster show. In many cases the viewer had to be positioned “just right” in order to prevent viewing distortions. Second, some paintings were placed at odd angles while other were placed too close together causing interference between groups viewing separate paintings. I brought these two things to them attention of one of the attendants and she said they were aware of these issues.

With my rant out of the way I can get back to the beauty of the worms at hand and how much I enjoyed viewing them. I will be writing more about individual pieces going forward.

A Framework for viewing the works of Frida Kahlo

I have visited a number of Frida Kahlo exhibits over the years with the most recent being the Virginia Museum of Fine Art's Frida: Beyond the Myth. One of the highlights of this exhibition was the display of a timeline highlighting the various periods of the artist's life.

I was able to use this construct to develop trees of the artist's work by period, using images captured during visits to said Virginia Museum, Frist Art Museum (Mexican Modernism), Brooklyn Museum (Frida Kahlo: Appearances can be Deceiving); and National Museum of Women in the Arts (Collection). Those trees are displayed in the charts below.

Some observations:

• A significant percentage of the paintings are portraits (13 of 31)
• There are many more drawings in the early portion of her career than later
• There is great consistency in depiction of her facial features in the portraits
• The first still life observed was Magnolias in 1945. She folllowed with a number of substantive works in that genre in later years
• The 1930s and 40s were her most productive years
• Surrealists wanted to claim her as their own and The Love Embrace of the Universe ..., provides ample evidence as to why.

Piero della Francesca’s Augustinian Altarpiece

 In 1454, Angelo di Giovanni di Simone d'Angelo commissioned a polyptych from Piero della Francesca for the High Altar of St Agostino in Borgo Sansepolchro. The polyptych was intended to fulfill the wishes of his late brother Simone and his widow Giovanna to procure spiritual benefits for the donor and his forbears.

The structural framework of the altarpiece had a history prior to Piero's storied work. The Franciscans had had the structure built in the early 1430s to support paintings by Antonio d'Anghari and Piero had worked with him to gesso the piece. The effort was abandoned after the Franciscans hired Sassetta to do the work instead.

In 1451 the Franciscans sold the framework to Angelo Giovanni di Simone. It was more than 40 years old when Piero began working on it as the mainstay of the Augustinian High Altar. The piece was completed in 1469 and served as the High Altar until it was disassembled in 1555 when the Augustinians moved. The 30 panels were dispersed, with eight currently housed in institutions in Europe and the US.

Previous attempts  have been made to unite the surviving pieces. The most successful, prior to the Poldi Pozzoli effort, was the Frick exhibition which displayed all of the pieces with the exception of the Poldi holding. When the Frick announced its pending closure for renovation, Poldi Pozzoli reached out to that institution and other holders with hosting requests. Those requests were granted and the pieces were reunited in an exhibition at Poldi Pozzi that ran from March 20 to June 24, 2024.

Piero della Francesca’s Baptism of Christ

 On my visit to the re-opening of the Frick Collection, I encountered their Piero della Francesca pieces and became excited because i will be doing something around this artist later in the year. Further investigation showed that the institution had hosted an exhibition of 7 of his pieces — six of which are based in the US — in 2013 and I dove into the archives to study the work done around that exhibition. This presentation of the artist’s Baptism of Christ is my first offering and draws heavily on the museum’s write-ups plus associated lectures.

The evolution of Large Language Models: From Rule-Based systems to ChatGPT

 Large language models have become a topic of immense interest and discussion in recent years. With the advent of advanced artificial intelligence technologies, we now have the ability to create machines that can process and understand human language at a level never before seen. One such technology that has gained widespread attention is ChatGPT, a large language model developed by OpenAI based on the GPT-3.5 architecture.

ChatGPT has been hailed for its ability to generate human-like text and perform a variety of language tasks with high accuracy. But how did we get here? In this foundational blog post, we will explore the history and underlying developments in large language models, tracing their origins from early rule-based systems to the deep learning-powered models of today. We will also examine the key breakthroughs and advancements that have led to the creation of ChatGPT and its predecessors, as well as the potential applications and ethical considerations of these technologies. So join us as we dive into the world of large language models and explore the possibilities they hold for the future of communication and AI.

The development of large language models has come a long way from early rule-based systems that relied on explicit rules to generate text or respond to queries. The advent of statistical language modeling techniques in the 1980s and 1990s led to the development of the Hidden Markov Model, which was used to model the probability of a sequence of words occurring in a given context.

In the 2000s, neural network-based language models began to emerge, which were able to learn more complex patterns and relationships between words than traditional statistical models. The introduction of the recurrent neural network (RNN) was a significant breakthrough in this area, as it was able to model sequential data such as language with great success. The long short-term memory (LSTM) and gated recurrent unit (GRU) architectures further improved the ability of RNNs to model long-term dependencies in language.

The advent of deep learning in the 2010s led to the development of even more powerful language models, including the Transformer architecture, which was introduced in 2017. The Transformer was able to learn even more complex relationships between words and achieve state-of-the-art results on a wide range of natural language processing tasks.

Today, large language models like ChatGPT have gained significant attention for their ability to generate human-like text and perform a variety of language tasks with high accuracy. These models are typically trained on massive datasets of text and use self-supervised learning techniques to learn representations of language that can be fine-tuned for specific tasks.

While the potential applications of large language models are vast, there are also ethical considerations and limitations to consider. For example, the potential for bias in training data and the impact of these models on the job market are just a few of the challenges that must be addressed as we continue to develop these technologies.

©EverythingElse238

Eighteenth and Nineteenth Century art movements: Rococo, Neoclassicism, and Romanticism

A graphical overview of the three most important art movements of the 18th and 19th centuries.

 

Neolithic-, Chalcolithic-, Bronze-, and Iron-Age civilizations and Art